CN114418699A - Product Recommended Methods, Apparatus, Equipment, Media and Program Products - Google Patents

Abstract

The disclosure provides a product recommendation method which can be applied to the technical field of big data or the financial field. The method comprises the following steps: obtaining a product recommendation strategy pool, wherein the product recommendation strategy pool comprises m first product recommendation strategies, and m is an integer greater than or equal to 2; polling and allocating users to m first product recommendation strategies based on a product recommendation strategy pool and a first traffic segmentation method; generating a product recommendation list based on a first product recommendation strategy distributed by a user and user access historical data, wherein the user access historical data comprises a user historical transaction product list and a user historical browsing product list; acquiring user access data, wherein the user access data is associated with a product recommendation list; and evaluating the m first product recommendation strategies based on a user access data set, wherein the user access data set comprises a set of all user access data in the test period. The present disclosure also provides a product recommendation apparatus, device, storage medium and program product.

Description

Product Recommended Methods, Apparatus, Equipment, Media and Program Products

technical field

The present disclosure relates to the technical field of big data or the financial field, and in particular to a product recommendation method, apparatus, device, medium and program product.

Background technique

At present, when user-based personalized recommendation for products sold online, the product recommendation strategy is formulated based on the customer's transaction behavior and browsing behavior, combined with certain time factors, such as the user's historical transaction time and historical browsing time. The formulated product recommendation strategy generates a product recommendation list, and then makes product recommendations based on the product recommendation list.

In the process of realizing the concept of the present disclosure, the inventor found that at least the following problems exist in the prior art:

Since the formulation method of product recommendation strategy is relatively simple and lacks a monitoring and evaluation mechanism, the effectiveness of the formulated product recommendation strategy needs to be improved.

SUMMARY OF THE INVENTION

In view of the above problems, the embodiments of the present disclosure provide a product recommendation method, apparatus, device, medium and program product for optimizing a product recommendation strategy and improving the effectiveness of product recommendation.

According to a first aspect of the present disclosure, a product recommendation method is provided, including: acquiring a product recommendation strategy pool, where the product recommendation strategy pool includes m first product recommendation strategies, where m is an integer greater than or equal to 2; The product recommendation strategy pool and the first traffic segmentation method allocate user polling to the m first product recommendation strategies; generate a product recommendation list based on the first product recommendation strategy assigned by the user and user access history data, wherein , the user access historical data includes the user's historical transaction product list and the user's historical browsing product list; obtain user access data, the user access data is associated with the product recommendation list; and based on the user access data set The m types of first products The recommendation strategy is evaluated, wherein the set of user access data includes the set of all user access data in the test period.

According to an embodiment of the present disclosure, the first product recommendation strategy is generated based on a product basic strategy and a first dynamic addition factor.

According to an embodiment of the present disclosure, the basic product strategy includes: calculating a basic product recommendation score for the products to be recommended included in the product list to be recommended, where the basic product recommendation score is based on a user transaction time factor and a user browsing time factor, The similarity of trading products and the similarity of browsing products are calculated.

According to an embodiment of the present disclosure, the user transaction time factor and the user browsing time factor are calculated based on a time factor algorithm, and the time factor algorithm includes: The historical product trading day is obtained based on the user's historical transaction product list; the browsing time interval between the historical product browsing day and the current visit day is calculated, and the historical product browsing day is obtained based on the user's historical browsing product list; based on the transaction time interval and a product term attribute to obtain the user transaction time factor; and to obtain the user browsing time factor based on the browsing time interval and the product term attribute.

According to an embodiment of the present disclosure, the method for generating the similarity of the trading product and the similarity of the browsing product includes: calculating the similarity between the product to be traded and the historical trading product of the user based on the product similarity index and a preset similarity index configuration , to obtain the similarity of the trading products; and calculate the similarity between the product to be traded and the historically browsed products of the user based on the product similarity index and the preset similarity index configuration, and obtain the similarity of the browsed products.

According to an embodiment of the present disclosure, the product similarity index includes two or more of product term, sales channel, risk level, minimum purchase amount, product type, transaction currency, and quota status.

According to an embodiment of the present disclosure, the first dynamic addition factor includes at least one popularity factor, and the popularity factor includes a star popularity factor, an age popularity factor, or a risk assessment popularity factor.

According to an embodiment of the present disclosure, the star popularity factor is determined based on historical access data of users of the same star rating; the age popularity factor is determined based on historical access data of users of the same age; the risk evaluation popularity factor is determined based on The historical access data of users with the same risk assessment level is determined, wherein the user star rating, the user age group and the user risk assessment level are determined based on preset assessment rules.

According to an embodiment of the present disclosure, generating the first product recommendation strategy based on the product basic strategy and the first dynamic addition factor includes: obtaining a first strategy weight vector, where the first strategy weight vector includes the product basic strategy weight and the first dynamic addition factor weights; and generating the first product recommendation strategy based on the product basic strategy, the first dynamic addition factor, and the first strategy weight vector.

According to an embodiment of the present disclosure, the evaluating the m types of first product recommendation policies based on the user access data set further includes: calculating a policy evaluation index value based on the user access data set, where the policy evaluation index value includes product basic policies The evaluation index value and the first product recommendation strategy evaluation index value; the first strategy effectiveness index value is calculated based on the product basic strategy evaluation index value and the first product recommendation strategy evaluation index value; and when the kth first product When the value of the first strategy effectiveness index corresponding to the recommended strategy is greater than or equal to a preset threshold, the kth first product recommendation strategy is marked as an effective strategy, where k satisfies 1≤k≤m and k is an integer.

According to an embodiment of the present disclosure, when the first strategy effectiveness index values corresponding to all the first product recommendation strategies are smaller than the preset threshold, the method further includes: executing the product recommendation strategy updating method j times until there is a second When the value of the second strategy effectiveness index corresponding to the product recommendation strategy is greater than or equal to a preset threshold, the second product recommendation strategy is marked as an effective strategy, where j is an integer greater than or equal to 1, wherein the product recommendation strategy The strategy updating method includes: acquiring at least one of a second strategy weight vector or a second dynamic addition factor, wherein the second strategy weight vector is obtained by adjusting the product basic strategy weight and the first dynamic addition factor in the first strategy weight vector. The relative proportion of the factor weight is obtained, and the second dynamic bonus factor is obtained by adjusting the type of popularity factor included in the first dynamic bonus factor; based on the second strategy weight vector and/or obtaining the second dynamic bonus factor, and the product base strategy to generate a second product recommendation strategy; and evaluating the second product recommendation strategy based on the same evaluation method as the first product recommendation strategy.

According to an embodiment of the present disclosure, when n types of effective strategies are included, the method further includes: sorting the n types of effective strategies according to the value of the first strategy effectiveness index in descending order to sort the first effective strategy is the optimal strategy, where n satisfies 2≤n≤m and n is an integer.

According to an embodiment of the present disclosure, when there are q effective strategies that are ranked first, where q satisfies 2≤q≤n and q is an integer, the method further includes: executing the traffic segmentation and updating method i times until When there is an effective strategy with the first and only ranking, the effective strategy is marked as the optimal strategy, where i is an integer greater than or equal to 1, wherein the traffic segmentation and update method includes: adjusting the first traffic segmentation method. obtain the second traffic segmentation method; allocate user polling to the q effective strategies based on the second traffic segmentation method; obtain the second strategy effectiveness index of the q effective strategies value, the second strategy effectiveness index value is calculated based on the same method as the first strategy effectiveness index value; the q effective strategies are sorted according to the second strategy effectiveness index value from large to small.

According to an embodiment of the present disclosure, after obtaining the optimal strategy, the method further includes: allocating all users to the optimal strategy; calculating the optimal strategy based on the same method as the first strategy effectiveness index value Effectiveness index value; the optimal strategy is evaluated based on the optimal strategy effectiveness index value and a preset evaluation period.

According to an embodiment of the present disclosure, the user access data includes page statistical data and access behavior data, and the page statistical data at least includes a touchpoint identifier, a product recommendation strategy, the amount of policy calls, and the number of product recommendations; the access behavior data is based on Obtain at least one of user transaction behavior and user browsing behavior.

According to an embodiment of the present disclosure, the strategy evaluation index includes a recommendation click rate and a purchase conversion rate.

A second aspect of the present disclosure provides a product recommendation device, including: a first acquisition module configured to acquire a product recommendation strategy pool, where the product recommendation strategy pool includes m types of first product recommendation strategies, where m is greater than or equal to 2 The first processing module is configured to allocate user polling to the m first product recommendation strategies based on the product recommendation policy pool and the first traffic segmentation method; the second processing module is configured to allocate user based The obtained first product recommendation strategy and user access historical data generate a product recommendation list, wherein the user access historical data includes the user's historical transaction product list and the user's historical browsing product list; the second acquisition module is configured to acquire user access data, The user access data is associated with a product recommendation list; and a third processing module is configured to evaluate the m types of first product recommendation strategies based on a user access data set, wherein the user access data set includes all the user access data sets in the test period. A collection of user access data.

A third aspect of the present disclosure provides an electronic device, comprising: one or more processors; a memory for storing one or more programs, wherein when the one or more programs are executed by the one or more programs When executed by the processor, one or more processors are caused to execute the above product recommendation method.

A fourth aspect of the present disclosure further provides a computer-readable storage medium having executable instructions stored thereon, the instructions, when executed by the processor, cause the processor to execute the above-mentioned product recommendation method.

A fifth aspect of the present disclosure also provides a computer program product, including a computer program, which implements the above-mentioned product recommendation method when the computer program is executed by a processor.

In the method provided by the embodiments of the present disclosure, by configuring multiple product recommendation strategies, segmenting user traffic, and allocating users to different product recommendation strategies according to the proportion of polling, it is possible to realize the selection of each product according to the actual operation effect. Compared with the single product recommendation strategy that lacks an evaluation mechanism, the evaluation of the recommendation strategy is conducive to improving the effectiveness of product recommendation, so as to improve the transaction rate in the actual sales of the product.

Description of drawings

The foregoing and other objects, features and advantages of the present disclosure will become more apparent from the following description of embodiments of the present disclosure with reference to the accompanying drawings, in which:

FIG. 1 schematically shows an application scenario diagram of a product recommendation method according to an embodiment of the present disclosure.

FIG. 2 schematically shows a flowchart of a product recommendation method according to an embodiment of the present disclosure.

FIG. 3 schematically shows a flowchart of a time factoring algorithm according to an embodiment of the present disclosure.

FIG. 4 schematically shows a flowchart of a method for generating the similarity of trading products and the similarity of browsing products according to an embodiment of the present disclosure.

5 schematically shows a flowchart of a method for generating a first product recommendation strategy based on a product basic strategy and a first dynamic addition factor according to an embodiment of the present disclosure.

FIG. 6 schematically shows a flowchart of a method for evaluating m types of first product recommendation strategies based on a user access data set according to an embodiment of the present disclosure.

FIG. 7 schematically shows a flowchart of a method for updating a product recommendation policy according to an embodiment of the present disclosure.

FIG. 8 schematically shows a flowchart of a method for screening an optimal strategy according to an embodiment of the present disclosure.

FIG. 9 schematically shows a flowchart of a method for screening an optimal strategy according to another embodiment of the present disclosure.

FIG. 10 schematically shows a flowchart of a method for updating traffic segmentation according to an embodiment of the present disclosure.

FIG. 11 schematically shows a flowchart of a method for evaluating an optimal strategy according to an embodiment of the present disclosure.

FIG. 12 schematically shows a structural block diagram of a product recommendation apparatus according to an embodiment of the present disclosure.

FIG. 13 schematically shows a block diagram of an electronic device suitable for implementing a product recommendation method according to an embodiment of the present disclosure.

Detailed ways

Hereinafter, embodiments of the present disclosure will be described with reference to the accompanying drawings. It should be understood, however, that these descriptions are exemplary only, and are not intended to limit the scope of the present disclosure. In the following detailed description, for convenience of explanation, numerous specific details are set forth in order to provide a thorough understanding of the embodiments of the present disclosure. It will be apparent, however, that one or more embodiments may be practiced without these specific details. Also, in the following description, descriptions of well-known structures and techniques are omitted to avoid unnecessarily obscuring the concepts of the present disclosure.

The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the present disclosure. The terms "comprising", "comprising" and the like as used herein indicate the presence of stated features, steps, operations and/or components, but do not preclude the presence or addition of one or more other features, steps, operations or components.

All terms (including technical and scientific terms) used herein have the meaning as commonly understood by one of ordinary skill in the art, unless otherwise defined. It should be noted that terms used herein should be construed to have meanings consistent with the context of the present specification and should not be construed in an idealized or overly rigid manner.

Where expressions like "at least one of A, B, and C, etc.," are used, they should generally be interpreted in accordance with the meaning of the expression as commonly understood by those skilled in the art (eg, "has A, B, and C") At least one of the "systems" shall include, but not be limited to, systems with A alone, B alone, C alone, A and B, A and C, B and C, and/or A, B, C, etc. ).

At present, when user-based personalized recommendation is made for products sold online, such as financial products sold by financial institution clients, it is mainly based on the customer's transaction behavior and browsing behavior, combined with certain time factors, such as the user's historical transaction time and history. At the browsing time, a product recommendation strategy is formulated, and a product recommendation list is generated based on the formulated product recommendation strategy, and then product recommendation is performed according to the product recommendation list. The formulation method of the above-mentioned product recommendation strategy is relatively simple and lacks a monitoring and evaluation mechanism, and the effectiveness and accuracy of the formulated product recommendation strategy need to be improved. On the other hand, how to dynamically monitor the effectiveness of product recommendation strategies, find the optimal strategy among a variety of feasible product recommendation strategies, and continuously evaluate the effectiveness of the optimal strategy is also an urgent problem to be solved.

In response to the above problems in the prior art, embodiments of the present disclosure provide a product recommendation method, including: acquiring a product recommendation strategy pool, where the product recommendation strategy pool includes m first product recommendation strategies, where m is greater than or equal to 2 is an integer; based on the product recommendation strategy pool and the first traffic segmentation method, user polling is allocated to the m types of first product recommendation strategies; based on the first product recommendation strategy assigned by the user and the user access historical data to generate a product A recommendation list, wherein the user access historical data includes a user historical transaction product list and a user historical browsing product list; obtain user access data, the user access data is associated with the product recommendation list; and based on the user access data set, the m A first product recommendation strategy is evaluated, wherein the user access data set includes the set of all user access data in the test period. By configuring multiple product recommendation strategies, segmenting user traffic, and assigning users to different product recommendation strategies according to the proportion of polling, the evaluation of each product recommendation strategy can be realized according to the actual operation effect, which is conducive to improving product quality. The effectiveness of the recommended strategy, thereby increasing the turnover rate in the actual sales of the product. Further, after evaluating the first product recommendation strategy, effective strategies can be screened out, and the effective strategies can be sorted to obtain the optimal strategy. Further, when no effective strategy is obtained, the product recommendation strategy update method can be executed to obtain the effective strategy; when there is more than one effective strategy ranked first, the traffic segmentation update method can be executed to obtain the optimal strategy. Furthermore, after obtaining the optimal strategy, continuous operation monitoring of the optimal strategy can be performed based on the full number of users to continuously evaluate the effectiveness of the optimal strategy.

It should be noted that the product recommendation method, apparatus, device, medium, and program product provided by the embodiments of the present disclosure can be used in big data technology in product recommendation-related aspects, and can also be used in various fields other than big data technology, such as financial field etc. The application fields of the product recommendation method, apparatus, device, medium, and program product provided by the embodiments of the present disclosure are not limited.

It should also be noted that in the technical solutions of the present disclosure, the collection, storage, use, processing, transmission, provision, disclosure and application of the user's personal information involved are all in compliance with the relevant laws and regulations, and the necessary measures have been taken. Confidentiality measures, and do not violate public order and good customs. In the technical solution of the present disclosure, the authorization or consent of the user is obtained before the user's personal information is obtained or collected.

The above operations to achieve at least one object of the present disclosure will be described below with reference to the accompanying drawings and the descriptions thereof.

FIG. 1 schematically shows an application scenario diagram of a product recommendation method according to an embodiment of the present disclosure.

As shown in FIG. 1 , an application scenario 100 according to this embodiment may include terminal devices 101 , 102 , and 103 , a network 104 and a server 105 . The network 104 is a medium used to provide a communication link between the terminal devices 101 , 102 , 103 and the server 105 . The network 104 may include various connection types, such as wired, wireless communication links, or fiber optic cables, among others.

The user can use the terminal devices 101, 102, 103 to interact with the server 105 through the network 104 to receive or send messages and the like. Various communication client applications may be installed on the terminal devices 101 , 102 and 103 , such as shopping applications, web browser applications, search applications, instant messaging tools, email clients, social platform software, etc. (only examples).

The terminal devices 101, 102, 103 may be various electronic devices having a display screen and supporting web browsing, including but not limited to smart phones, tablet computers, laptop computers, desktop computers, and the like.

The server 105 may be a server that provides various services, such as a background management server (just an example) that provides support for websites browsed by users using the terminal devices 101 , 102 , and 103 . The background management server can analyze and process the received user requests and other data, and feed back the processing results (such as web pages, information, or data obtained or generated according to user requests) to the terminal device.

It should be noted that the product recommendation method provided by the embodiment of the present disclosure may generally be executed by the server 105 . Correspondingly, the product recommendation apparatus provided by the embodiment of the present disclosure may generally be provided in the server 105 . The product recommendation method provided by the embodiment of the present disclosure may also be executed by a server or server cluster that is different from the server 105 and can communicate with the terminal devices 101 , 102 , 103 and/or the server 105 . Correspondingly, the product recommendation apparatus provided by the embodiments of the present disclosure may also be provided in a server or server cluster that is different from the server 105 and can communicate with the terminal devices 101 , 102 , 103 and/or the server 105 .

It should be understood that the numbers of terminal devices, networks and servers in FIG. 1 are merely illustrative. There can be any number of terminal devices, networks and servers according to implementation needs.

Based on the scenario described in FIG. 1 , the product recommendation method of the disclosed embodiment will be described in detail below with reference to FIGS. 2 to 11 .

FIG. 2 schematically shows a flowchart of a product recommendation method according to an embodiment of the present disclosure.

As shown in FIG. 2 , the product recommendation method of this embodiment includes operations S210 to S250.

In operation S210, a product recommendation policy pool is obtained.

According to an embodiment of the present disclosure, the product recommendation policy pool may be preconfigured. It can be understood that the product recommendation strategy pool may include m types of first product recommendation strategies, where m is an integer greater than or equal to 2. The first product recommendation strategy may be an unassessed product recommendation strategy. The first product recommendation strategy may be evaluated in the subsequent evaluation process to facilitate adjustment and optimization of the strategy.

In operation S220, user polling is allocated to the m first product recommendation strategies based on the product recommendation strategy pool and the first traffic segmentation method.

According to the embodiments of the present disclosure, the traffic segmentation method can be used to allocate the user polling routes to the m types of first product recommendation strategies according to the preset traffic allocation ratio. Wherein, the first traffic segmentation method may be a traffic segmentation method with a preset traffic distribution ratio, and a typical first traffic segmentation method may include average traffic segmentation, that is, users are averaged according to the number of types of the first product recommendation strategy. Routing into each category, so as to balance the effectiveness of each first product recommendation strategy without considering the traffic factor.

In some specific embodiments, it is assumed that there are four first product recommendation strategies in the product recommendation strategy pool, and the sum of the traffic distribution ratios of the four product recommendation strategies is 100%. According to user access conditions, traffic can be allocated in a round-robin manner. For example, the first user to access is routed to product recommendation policy Y0, the second user to access product recommendation policy Y1, and the third user to access product recommendation policy Y1. Product recommendation strategy Y ₃ , the fourth visited user is routed to the product recommendation strategy Y4, the fifth visited user is routed to the product recommendation strategy Y0, and so on, and polled in order to ensure that all visited users are routed to the product recommendation strategy Y0. The proportions of the above product recommendation strategies are the same, that is, the traffic allocation for each of the product recommendation strategies Y0, Y1, _Y2 , and _Y3 is 25%. Therefore, the influence of user traffic can be excluded, and the effectiveness of each first product recommendation strategy can be evaluated under the same proportion of user traffic.

In operation S230, a product recommendation list is generated based on the first product recommendation policy assigned by the user and user access history data.

According to the embodiments of the present disclosure, it can be understood that after the user logs in to the product recommendation platform, user access history data can be obtained based on the user ID, and the user access history data can at least include the user's historical transaction product list and the user's historical browsing product list. In order to evaluate the first product recommendation strategy, after the first product recommendation strategy is configured, a product recommendation list for the current user may be generated based on the first product recommendation strategy actually assigned by the user and user access history data. It can be understood that one or more recommended products may be included in the product recommendation list, which may be products on sale, wherein the products on sale may be determined based on the current user access time point. The user's historical transaction or browsing information can be obtained through the user's access historical data, so as to obtain the actual recommended products in combination with the product recommendation strategy.

In operation S240, user access data is obtained, where the user access data is associated with the product recommendation list.

In operation S250, the m first product recommendation strategies are evaluated based on the user access data set.

According to the embodiments of the present disclosure, in the process of the user accessing the product recommendation platform, the user access data can be acquired in real time, and further, the set of all user access data in the test period, that is, the user access data set can be acquired.

It can be understood that the user access data may be acquired based on the tracking technology, and the tracking may be set based on user touch points. Among them, the meaning of the touch point is the user's vision, hearing, smell, feeling, etc., and the product recommendation platform, such as the relevant channels of financial institutions, more specifically, such as financial institutions' outlets, staff, smart terminals and online channels between the points that can be touched. Embodiments of the present disclosure can implement product recommendation through touchpoint marketing. The meaning of touchpoint marketing is to create and package existing or potential touchpoints to make users pay attention and feel the value of the products and services provided by the product recommendation platform, so as to realize the user's sense of identity with the product recommendation platform. Contacts can generally include on-line contacts and off-line contacts. The contacts involved in the embodiments of the present disclosure may include online contacts, for example, may include a mobile banking transaction page, a menu, and a login IP. Understandably, users can access product recommendation lists through touchpoints. In the embodiment of the present disclosure, the touch point may be set on a page frequently visited by the user. In some specific embodiments, in the business scenario of mobile banking, the touch points can be set on the page of successful purchase of wealth management products, the page of wealth management product recommendation on the monthly bill, the page of my assets on the monthly bill, the page of guess your favorite wealth management product, and the wealth management product page. Product expiration reminder page, etc. Users can browse recommended wealth management products on the contact page, click on wealth management products to view detailed information, or purchase wealth management products. It can be understood that after the user accesses the contact point, the system can record the user information (including but not limited to the customer number), and record the contact point identifier and contact point name for the user to guard against.

According to an embodiment of the present disclosure, buried points may be set based on user touch points. The meaning of burying is to collect some information in a specific process in the application to track the usage of the application, and then use it to further optimize the product or provide data support for operations. In the embodiment of the present disclosure, the user access data collected according to the buried points includes page statistics data and access behavior data. Among them, the page statistics at least include touch point identifiers, product recommendation strategies, strategy calls, and product recommendations. The access behavior data is obtained based on at least one of user transaction behavior and user browsing behavior, and the access behavior data may include page stay time, page bounce rate, recommended product clicks, recommended product purchases, and the like. In addition, information such as the type of information material and the name of the calling interface can also be collected through buried points.

In this way, the overall invocation of each product recommendation strategy in the test period can be counted based on the user access data set, and the strategy invocation can be converted into the actual user click rate and purchase rate, so as to realize the recommendation of each first product. Evaluation of the strategy.

In the embodiment of the present disclosure, by configuring multiple product recommendation strategies, segmenting user traffic, and allocating users to different product recommendation strategies according to the proportion of polling, it is possible to realize the optimization of each product recommendation strategy according to the actual operation effect. Evaluation is beneficial to improve the effectiveness of the product recommendation strategy, thereby increasing the transaction rate in the actual sales of the product.

According to an embodiment of the present disclosure, the first product recommendation strategy is generated based on a product basic strategy and a first dynamic addition factor.

In the embodiment of the present disclosure, the product-based strategy may mainly consider the customer's own transaction behavior, that is, internal influencing factors. According to the customer's transaction behavior and browsing behavior, combined with certain time factors (such as the user's historical transaction time and historical browsing time), and the consideration of the similarity of the product to be recommended with the user's historical transaction product or user's historical browsing product.

Specifically, the basic product strategy may include: calculating the basic product recommendation score for the products to be recommended included in the list of products to be recommended, wherein the basic product recommendation score is based on a user transaction time factor, a user browsing time factor, and similar transaction products. degree and browsing product similarity are calculated. Wherein, the user transaction time factor and the user browsing time factor are calculated based on the time factor algorithm. The time factor can be determined according to the length of the interval between the trading day when the customer once purchased the product or the browsing day of the product and the current customer visiting the page. The specific time factor value can be set according to the length of the interval time. For example, the shorter the interval length, the larger the time factor. The similarity of trading products or browsing products is the similarity between the products to be recommended and the historical trading products or historical browsing products of the user. Therefore, after the user accesses the product recommendation system, the user can query the user's historical transaction product list and the user's historical browsing list based on the user's anti-interrogation data, and determine whether there is transaction information and browsing information. The time factor for the user's historically traded products and/or the user's historically browsed products can then be calculated. Further, the similarity between the user's historical transaction products or the user's historical browsing products and the products in the product list to be recommended can be calculated, based on the user's transaction time factor, the user's browsing time factor, the transaction product similarity and the browsing product similarity. Recommended list.

FIG. 3 schematically shows a flowchart of a time factoring algorithm according to an embodiment of the present disclosure.

As shown in FIG. 3 , the time factor algorithm of this embodiment includes operations S310 to S340.

In operation S310, the transaction time interval between the historical product transaction day and the current access day is calculated.

According to an embodiment of the present disclosure, the historical product trading day is obtained based on the user's historical trading product list.

In operation S320, the browsing time interval between the historical product browsing day and the current visiting day is calculated.

According to an embodiment of the present disclosure, the historical product browsing day is obtained based on the user's historical browsing product list.

In operation S330, the user transaction time factor is obtained based on the transaction time interval and the product term attribute.

In operation S340, the user browsing time factor is obtained based on the browsing time interval and the product term attribute.

In some specific embodiments, the products to be recommended may be financial products, such as wealth management dishes. The product similarity index may include two or more of product term, sales channel, risk level, minimum purchase amount, product type, transaction currency, and quota status.

According to a specific embodiment of the present disclosure, the user transaction time factor and the user browsing time factor can be configured based on the term attribute of the financial product.

In a typical example, the term attribute of financial products can be divided into live money management, 0-6 months, 6-12 months, and more than 12 months.

For the user transaction time factor, based on the term attribute of the financial product, it can be configured based on the following assignments:

For wealth management products whose term attribute is live money management (can be subscribed and redeemed at any time), if the transaction time interval between the historical product transaction day and the current access day is 0-1 month, the user transaction time factor can be assigned a value of 100. Similarly, assign 90 for trade interval 1-3 months, assign 80 for trade interval 3-6 months, assign 70 for trade interval 6-12 months, and trade interval 12 months - 60 for 24 months, and 40 for transactions over 24 months.

For wealth management products whose duration attribute is 0-6 months, and the transaction time interval between the historical product transaction day and the current access day is 0-6 months, the user's transaction time factor can be assigned a value of 100, and the transaction time interval is 6-12 The monthly value is 80, the transaction time interval is 12 months to 24 months, the value is 60, and the transaction time interval is more than 24 months.

For wealth management products whose duration attribute is 6-12 months, and the trading time interval between the trading day of the historical product and the current access day is 0-12 months, the time factor of the trading product can be assigned a value of 100, and the time period is 12 months-24 Monthly assignment is 80, and time is 40 for more than 24 months.

For wealth management products whose duration attribute is more than 12 months, and the transaction time interval between the historical product transaction day and the current access day is 0-12 months, the user's transaction time factor can be assigned 100, and the transaction time interval is 12 months-24 The monthly value is 80, the transaction interval is 24 months to 36 months, and the transaction interval is 60, and the transaction interval is more than 36 months.

As for the user browsing time factor, it can be understood that since browsing a product to a certain extent only expresses interest in the product and does not necessarily buy it, it can be calculated by multiplying the result by 50% according to the same calculation method as the transaction product time factor. Specifically, based on the term attribute of the financial product, it can be configured based on the following assignments:

For wealth management products whose attribute is live money management (can be subscribed and redeemed at any time), if the browsing time interval between the historical product browsing day and the current visiting day is 0-1 month, the user browsing time factor can be assigned a value of 50. Browsing the assignment 45 with a time interval of 1-3 months. If the browsing time interval is 3-6 months, assign 40, if the browsing time interval is 6-12 months, assign 35, if the browsing time interval is 12-24 months, assign 30, and if the browsing time interval is more than 24 months Assign the value 20.

For wealth management products whose attributes are 0-6 months, and the browsing time interval between the historical product browsing day and the current visiting day is 0-6 months, the user browsing time factor can be assigned a value of 50, and the browsing time interval is 6-12 months The assignment is 40, the browsing interval is 12 months to 24 months, the assignment is 30, and the browsing interval is 24 months or more, the assignment is 20.

For wealth management products whose attributes are 6-12 months, and the browsing time interval between the historical product browsing day and the current visiting day is 0-12 months, the user browsing time factor can be assigned a value of 50, and the browsing time interval is 12 months-24 40 is assigned to the month, and 20 is assigned to the browsing interval of more than 24 months.

For wealth management products whose attributes are more than 12 months, and the browsing time interval between the historical product browsing day and the current visiting day is 0-12 months, the user browsing time factor can be assigned a value of 50. A value of 40 is assigned to a browsing interval of 12 months to 24 months, a value of 30 is assigned to a browsing interval of 24 months to 36 months, and a value of 20 is assigned to a browsing interval of more than 36 months.

FIG. 4 schematically shows a flowchart of a method for generating the similarity of trading products and the similarity of browsing products according to an embodiment of the present disclosure.

As shown in FIG. 4 , the method for generating the similarity of trading products and the similarity of browsing products in this embodiment includes operations S410 to S420 .

In operation S410, the similarity between the product to be recommended and the user's historical transaction products is calculated based on the product similarity index and the preset similarity index configuration, and the similarity of the transaction products is obtained.

In operation S420, the similarity between the product to be recommended and the user's historical browsing products is calculated based on the product similarity index and the preset similarity index configuration, and the browsing product similarity is obtained.

According to an embodiment of the present disclosure, the product similarity may have multiple influencing factors. Correspondingly, the relevant influencing factors can be set as the product similarity index. In some specific embodiments, the preset configuration can be performed by directly assigning the similarity index. When multiple similarity indexes are included, joint evaluation can also be performed based on a preset calculation method. A typical calculation method can be It includes the calculation and summation after configuring the weight of each product similarity index, and it is also possible to assign and multiply multiple similarity indexes.

In some specific embodiments, online financial product recommendation by a financial institution is used as an example of an application scenario. The product similarity index may include two or more of product term, sales channel, risk level, minimum purchase amount, product type, transaction currency, and quota status.

In a specific example, the product term may include live money management, 0-6 months, 6-12 months, and more than 12 months. The sales channel can include self-operated sales or agency sales. Correspondingly, you can set the self-operated/resourced sales labels to include self-operated and agency sales. The minimum purchase amount can include 0-10,000, 1-50,000, 5-500,000, 50-5 million, and more than 5 million. Risk levels can include low risk, low risk, moderate risk, higher risk, and high risk. Product types can include fixed income, equity, commodities (such as precious metals, etc.), financial derivatives, and hybrids. Currency can include RMB, USD, GBP, EUR. The quota status can include remaining quota and currently no quota. Based on the above product similarity index, the preset similarity index configuration may include: 1) If the product to be recommended is consistent with the product period of the user's historical transaction product (or user's historical browsing product), the period similarity is taken as 1; the product period If they are inconsistent, the similarity of the period is taken as 0. 2) If the sales channel of the product to be recommended is the same as that of the user’s historical transaction product (or user’s historical browsing product), the similarity of the sales channel is taken as 1; if the sales channel is inconsistent, the sales channel is taken as the same. The similarity is 0.5. 3) If the minimum purchase amount of the product to be recommended is different from the minimum purchase amount of the user's historical transaction product (or user's historical browsing product), and the minimum purchase amount of the two is in the adjacent interval, configure the minimum purchase amount similarity is 0.75; if the minimum purchase amount of the product to be recommended is exactly the same as the minimum purchase amount of the user's historical transaction product (or user's historical browsing product), the similarity of the configured minimum purchase amount is 1; in other cases, the configured minimum purchase amount similarity is 0.4) If the risk level of the product to be recommended and the user's historical transaction product (or user's historical browsing product) are in the adjacent range, configure the risk level similarity to 0.80; If the product) has the same risk level, the risk level similarity is 1; in other cases, the configuration risk level similarity is 0. 5) If the product to be recommended is the same as the user's historical transaction product (or user's historical browsing product) product type, then The product type similarity is 1, otherwise it is 0. 6) If the product to be recommended is consistent with the transaction currency of the user's historical transaction product (or user's historical browsing product), the transaction currency similarity is 1, otherwise it is 0. 7) If If the quota status of the product to be recommended is still available, the similarity of the quota status is 1; if the quota status of the product to be recommended is no quota, the similarity of the quota status is 0.

Thus, the similarity between the product to be recommended and the user's historical trading products can be calculated based on the product similarity index and the preset similarity index configuration, respectively, to obtain the similarity of the trading products; and based on the product similarity index and the preset similarity index The configuration calculates the similarity between the product to be recommended and the user's historical browsing products, and obtains the similarity of the browsing products. In an exemplary product similarity calculation method, whether it is trading product similarity or browsing product similarity, it can be calculated by multiplying the preset assignments of the product similarity index configuration, that is, product similarity = product duration Similarity × sales channel similarity × minimum purchase amount similarity × risk level similarity × product type similarity × transaction currency similarity × quota status similarity. In another exemplary method for calculating the similarity of products, the calculation may also be performed by dividing the sum of the similarity indices of each product by the sum total when the similarity indices of each product are all 1. It can be understood that the above assignment and calculation methods are only illustrative examples, and the assignment methods and calculation methods may be adjusted based on actual operation conditions. It can be understood that if the product to be recommended is a product that has been traded before, for example, the product identifiers are the same, there is no need to calculate the similarity of the traded product, and the similarity of the traded product is directly determined to be 1. If the product to be recommended is a product that has been browsed, such as a product If the identifiers are the same, there is no need to calculate the similarity of the browsing products, and the similarity of the browsing products is directly determined to be 1. When the product to be recommended is neither a product that has been traded nor browsed, the similarity of the traded product and the similarity of the browsed product can be obtained according to the index configuration and calculation method listed in the embodiment of the present disclosure.

According to an embodiment of the present disclosure, after obtaining the user transaction time factor, user browsing time factor, transaction product similarity, and browsing product similarity, a product basic recommendation score can be comprehensively calculated for each product to be recommended according to the product identifier. For example, the user transaction time factor, user browsing time factor, transaction product similarity, and browsing product similarity are respectively given a certain weight and then multiplied or added together to calculate the product basic recommendation score.

In an example, the product-based recommendation strategy Y ₀ can calculate the product-based recommendation score ∑Y ₀ by formula (1):

∑Y ₀ =a×c _a +b×c _b Formula (1)

Among them, a is the user transaction time factor, b is the user browsing time factor, c _a is the transaction product similarity, and c _b is the browsing product similarity.

It should be noted that when there are multiple user historical transaction products and/or multiple user historical browsing products, the time factor of each user's transaction product × transaction product similarity, and the user browsing time factor × browsing product similarity can be calculated separately. , add up and take the average value to get the product basic recommendation score.

After obtaining the basic recommendation scores of each product to be recommended, the products to be recommended may be ranked based on the scores. When a user accesses a touchpoint, based on the configuration of the corresponding touchpoint page, a specified number of products before the total ranking can be displayed as recommended products, and the above-mentioned recommendation method can be used as a basic product recommendation strategy Y ₀ .

According to an embodiment of the present disclosure, the first dynamic addition factor introduces an external reference factor, and the external reference factor is some attribute dimension that is associated with the purchase behavior, which can be obtained through products of other users that are the same as the user in the relevant attribute dimension. Transaction preferences to predict users' transaction and browsing behavior. Taking the online wealth management product sales of financial institutions as an example, when users purchase wealth management products, their risk appetite and asset allocation needs for wealth management products are directly related to the user's income or asset level. Considering that the user has an understanding of the same asset level, or The purchase preferences of other users of the same age group or the same risk assessment level, the first dynamic bonus factor can include the popularity factor, and the popularity factor can be the same asset level, or the same age group, or the same risk assessment level. The popularity of users buying products.

In an embodiment of the present disclosure, the first dynamic addition factor includes at least one popularity factor, and the popularity factor includes a star popularity factor, an age popularity factor, or a risk assessment popularity factor. According to an embodiment of the present disclosure, the popularity factor may be determined based on user historical access data. Specifically, the star popularity factor may be determined based on historical access data of users of the same star rating; the age popularity factor may be determined based on historical access data of users of the same age; the risk assessment popularity factor may be determined based on the same The risk assessment level is determined based on the historical access data of the user, wherein the user star rating, the user age group and the user risk assessment level are determined based on preset assessment rules.

In an example, taking a bank as an example, the user star rating is mainly related to the average daily financial asset size of the user in the bank for the previous month. For example, at the beginning of each calendar month, the bank can assess the star rating based on the average daily financial asset size of the user in the previous month. New users will have their first star rating at the beginning of the next month after opening an account. The financial assets included in the star rating range, including the user's local and foreign currency deposits in the bank, as well as the total market value of wealth management, funds, insurance, government bonds, third-party custody, and account transaction assets. For example, individual customer star ratings from high to low can include private banking, seven-star, six-star, five-star, four-star, three-star, two-star, one-star, and each star is related to the previous month. The scale of monthly average financial assets corresponds, for example, there is the following correspondence: Private bank level: monthly average financial assets of 8 million yuan (inclusive) or more; seven-star: monthly average financial assets of 6 million yuan (inclusive) to 8 million yuan (exclusive); six-star: monthly average financial assets of 1 million yuan (inclusive) - 6 million yuan (exclusive); five-star: monthly average financial assets of 200,000 yuan (inclusive) - 1 million yuan (not inclusive) Inclusive); four-star: monthly average financial assets of 50,000 yuan (inclusive) to 200,000 yuan (exclusive); three-star: monthly average of financial assets of 10,000 yuan (inclusive) to 50,000 yuan (exclusive) ; Two-star: The monthly average daily financial asset is less than 10,000 yuan (excluding zero balance); One-star: The monthly average financial asset balance is zero. You can obtain the sales ranking information of wealth management products purchased by users with the same star rating in the last month, and obtain the star popularity factor of the same user star rating according to the sales ranking. In some typical examples, the specific value of the star popularity factor can be marked by means of assignment. For example, you can take the top 20 wealth management products according to the sales ranking of wealth management products purchased by the same user star, and the star popularity factor can be 100, 95, 90, 85, 80, 75, 70, 65, 60, 55, 50, 45, 40, 35, 30, 25, 20, 15, 10, 5, the star popularity factor of other financial products is 0.

Similarly, the age popularity factor purchased by customers of the same age group can be assigned: according to the sales ranking of wealth management products purchased by users of the same age group, the top 20 wealth management products are selected, and their age popularity factors are 100 and 95. , 90, 85, 80, 75, 70, 65, 60, 55, 50, 45, 40, 35, 30, 25, 20, 15, 10, 5, and the age popularity factor of other financial products is 0. Among them, the age groups are divided as follows: youth: 18-35 years old; young and middle-aged: 36-50 years old; middle-aged: 51-65 years old; old age: 65 years old and above.

Similarly, according to the sales ranking of wealth management products purchased by other users with the same risk assessment rating, the top 20 wealth management products can be selected, and the risk assessment popularity factor can be 100, 95, 90, 85, 80, 75, 70, 65, 60, 55, 50, 45, 40, 35, 30, 25, 20, 15, 10, 5, and the popularity factor for risk assessment of other financial products is 0.

Thus, a variety of first dynamic addition factors can be constructed based on the selection and combination of popularity factors, thereby obtaining multiple first product recommendation strategies, and evaluating each first product recommendation strategy in actual operation.

It can be understood that the above-mentioned popularity factor can be adjusted according to the actual operation process, and the adjustment includes adding another attribute of the popularity factor associated with the user's product purchase preference, so as to optimize the construction of the first dynamic bonus factor. It is worth noting that, in the embodiment of the present disclosure, the popularity factor may be associated with user information. When it comes to obtaining user information, such as before obtaining the user's information, the user's consent or authorization can be obtained. For example, before obtaining the age popularity factor, a request for obtaining user information may be sent to the user. Under the condition that the user agrees or authorizes that the user information can be obtained, the step of obtaining the age popularity factor is performed.

5 schematically shows a flowchart of a method for generating a first product recommendation strategy based on a product basic strategy and a first dynamic addition factor according to an embodiment of the present disclosure.

As shown in FIG. 5 , the method for generating the first product recommendation strategy based on the product basic strategy and the first dynamic addition factor in this embodiment includes operations S510 to S520.

In operation S510, a first strategy weight vector is obtained, where the first strategy weight vector includes the product basic strategy weight and the first dynamic addition factor weight.

In operation S520, the first product recommendation strategy is generated based on the product basic strategy, the first dynamic addition factor, and the first strategy weight vector.

According to an embodiment of the present disclosure, when generating the first product recommendation strategy, a first strategy weight vector may be set, wherein the first strategy weight vector includes a product basic strategy weight value and a first dynamic addition strategy weight value. Thus, the product basic strategy score can be obtained through the product basic recommendation score and the product basic strategy weight value; the dynamic additional factor score can be obtained through the first dynamic additional factor and the first dynamic additional strategy weight value. Therefore, after obtaining the product identification to be recommended, you can combine the product basic strategy score and dynamic bonus factor score to calculate the recommended total score of each product to be recommended, and then sort according to this, and display the specified number of products ranked at the top as Recommend products as the first product recommendation strategy. In a specific example, the top 2 or the top 10 products in the total ranking may be displayed as recommended products according to the settings of the user visiting the touchpoint page.

In one example, the first product recommendation strategy Y ₁ can calculate the first product recommendation score ∑Y ₁ by formula (2):

∑Y ₁ =(a×c _a +b×c _b )*λ ₁ +d*λ ₂ Formula (2)

where a is the user transaction time factor, b is the user browsing time factor, c _a is the similarity of trading products, c _b is the similarity of browsing products, λ ₁ is the weight value of the basic product strategy, d is the star popularity factor assignment, λ ₂ is the first dynamic addition factor weight. In one example, the importance of the product base strategy and the first dynamic bonus factor can be considered equally, and λ ₁ =λ ₂ =50% is set.

After obtaining the first product recommendation score ∑Y ₁ of each product to be recommended, the products to be recommended may be ranked based on the scores. When a user accesses a touchpoint, based on the configuration of the corresponding touchpoint page, a specified number of products before the total ranking can be displayed as recommended products, and the above-mentioned recommendation method can be used as the first product recommendation strategy Y ₁ .

Similarly, the first product recommendation strategy Y2 can calculate the first product recommendation score ∑Y2 by formula (2):

∑Y ₂ =(a×c _a +b×c _b )*λ ₁ +e*λ ₃ Formula (2)

where a is the user’s transaction time factor, b is the user’s browsing time factor, c _a is the similarity of trading products, c _b is the similarity of browsing products, λ ₁ is the weight value of the product’s basic strategy, e is the star popularity factor assignment, λ ₃ is the first dynamic addition factor weight. In one example, the importance of the product base strategy and the first dynamic bonus factor can be considered equally, and λ ₁ =λ ₃ =50% is set.

After obtaining the first product recommendation score ∑Y ₂ of each product to be recommended, the products to be recommended may be ranked based on the scores. When a user accesses a touchpoint, based on the configuration of the corresponding touchpoint page, a specified number of products before the total ranking can be displayed as recommended products, and the above-mentioned recommendation method can be used as the first product recommendation strategy Y ₂ .

Similarly, the first product recommendation strategy Y ₃ can calculate the first product recommendation score ∑Y ₃ by formula (3):

∑Y ₃ =(a×c _a +b×c _b )*λ ₁ +(d+e)*λ ₄ Formula (3)

Among them, a is the user transaction time factor, b is the user browsing time factor, c _a is the similarity of trading products, c _b is the similarity of browsing products, input ₄ is the weight value of the basic product strategy, e is the star popularity factor assignment, and input ₂ is the first dynamic addition factor weight. In one example, the importance of the product base strategy and the first dynamic bonus factor can be considered to the same degree, and λ ₁ =λ ₄ =50% is set.

After obtaining the first product recommendation score ∑Y ₃ of each product to be recommended, the products to be recommended may be ranked based on the scores. When a user accesses a touchpoint, based on the configuration of the corresponding touchpoint page, a specified number of products before the total ranking can be displayed as recommended products, and the above-mentioned recommendation method can be used as the first product recommendation strategy Y ₃ .

It can be understood that after the first dynamic addition factor introduces more popularity factors, the first product recommendation strategy that introduces new popularity factors can be generated according to a method similar to the first product recommendation strategy Y ₁ , Y2 , and Y3 .

FIG. 6 schematically shows a flowchart of a method for evaluating m types of first product recommendation strategies based on a user access data set according to an embodiment of the present disclosure.

As shown in FIG. 6 , the product recommendation method of this embodiment includes operations S610 to S640, or operations S610 to S630, and operations S650.

In operation S610, a policy evaluation index value is calculated based on the user access data set.

In operation S620, a first strategy effectiveness index value is calculated based on the product basic strategy evaluation index value and the first product recommendation strategy evaluation index value.

In operation S630, it is determined whether the first strategy effectiveness index value corresponding to the kth first product recommendation strategy is greater than or equal to a preset threshold, where k satisfies 1≤k≤m and k is an integer.

When the value of the first strategy effectiveness index corresponding to the kth first product recommendation strategy is greater than or equal to the preset threshold, operation S640 is performed.

In operation S640, the k-th first product recommendation strategy is marked as an effective strategy.

According to an embodiment of the present disclosure, the strategy evaluation index value includes a product basic strategy evaluation index value and a first product recommendation strategy evaluation index value. It can be understood that the first product recommendation strategy should have higher effectiveness than the product basic strategy. In this way, the product basic strategy evaluation index value can be calculated based on the user access data set, and then the first strategy effectiveness index value can be calculated by comparing the product basic strategy evaluation index value and the first product recommendation strategy evaluation index value to indicate the introduction of the first dynamic After the factor, the degree of improvement of the effectiveness of the product recommendation strategy. For example, whether the corresponding first product recommendation strategy is an effective strategy may be determined by whether the ratio of the first product recommendation strategy evaluation index value to the product basic strategy evaluation index value is greater than a preset threshold. In one example, the preset threshold value can be 150%, that is, when the ratio of the first product recommendation strategy evaluation index value corresponding to a certain first product recommendation strategy to the product basic strategy evaluation index value is greater than or equal to 150%, mark the The first product recommendation strategy is an effective strategy.

According to an embodiment of the present disclosure, the strategy evaluation indicators may include a recommendation click rate and a purchase conversion rate. The recommended click-through rate may be calculated based on the number of clicks on the recommended product/the number of recommendations in the user access data. The purchase conversion rate may be calculated based on the number of purchases/recommendations of the recommended products in the user access data. It can be understood that when conducting strategy evaluation, auxiliary reference indicators, such as transaction amount, can also be introduced to evaluate product solutions from multiple dimensions, so as to realize the screening of effective strategies and even optimal strategies.

According to an embodiment of the present disclosure, when the first strategy effectiveness index values corresponding to all the first product recommendation strategies are smaller than the preset threshold, operation S650 is performed.

In operation S650, the product recommendation strategy updating method is executed j times, and when there is a second strategy effectiveness index value corresponding to the second product recommendation strategy greater than or equal to a preset threshold, the second product recommendation strategy is marked as an effective strategy, Among them, j is an integer greater than or equal to 1.

According to the embodiment of the present disclosure, when the first strategy effectiveness index values corresponding to all the first product recommendation strategies are all smaller than the preset threshold, the second strategy effectiveness index value can be obtained by updating the product recommendation strategy to improve the product recommendation strategy Evaluate again. The updating process may be performed cyclically until a second policy effectiveness index value greater than or equal to a preset threshold is obtained.

FIG. 7 schematically shows a flowchart of a method for updating a product recommendation policy according to an embodiment of the present disclosure.

As shown in FIG. 7 , the method for updating a product recommendation policy of this embodiment includes operations S710 to S730.

In operation S710, at least one of a second policy weight vector or a second dynamic addition factor is obtained.

In operation S720, a second product recommendation strategy is generated based on the second strategy weight vector and/or the obtained second dynamic addition factor, and the product basic strategy.

In operation S730, the second product recommendation strategy is evaluated based on the same evaluation method as the first product recommendation strategy.

According to the embodiment of the present disclosure, the first product recommendation strategy can be adjusted by adjusting the specific content of the strategy weight vector or the dynamic addition factor. Wherein, the second strategy weight vector can be obtained by adjusting the relative ratio of the product basic strategy weight and the first dynamic addition factor weight in the first strategy weight vector. On the other hand, the second dynamic addition factor may be obtained by adjusting the type of popularity factor included in the first dynamic addition factor. In an example, in the first strategy weight vector, the product basic strategy weight and the first dynamic addition factor weight are both 50%, then in the first strategy weight vector, the actual situation can be evaluated based on the operation situation, and the product The weight of the basic strategy is adjusted to 30%, and the weight of the first dynamic bonus factor is all 70%. In another example, adjusting the types of popularity factors included in the first dynamic bonus factor may include new types of popularity factors, such as newly-added income popularity factors, that is, popularity factors for users in the income range of the same year to purchase wealth management products, Its value method can be similar to other popularity factors. Thus, the second product recommendation strategy can be generated based on the second strategy weight vector and/or the second dynamic addition factor, and the product basic strategy. Further, the method for generating the second product recommendation strategy may be the same as the method for generating the first product recommendation strategy, and details are not described herein again. It can be understood that, after obtaining the generation method of the second product recommendation strategy, the second product recommendation strategy can be evaluated based on the same evaluation method as the first product recommendation strategy.

According to an embodiment of the present disclosure, an effective policy may include more than one. In order to maximize the function of the product recommendation strategy, the embodiments of the present disclosure include the step of screening the optimal strategy. It can be understood that when there is only one effective strategy obtained through evaluation, the effective strategy is taken as the optimal strategy. Further, when there are n kinds of effective strategies, where n satisfies 2≤n≤m and n is an integer, a method of screening the optimal strategy from the n kinds of effective strategies is also included.

FIG. 8 schematically shows a flowchart of a method for screening an optimal strategy according to an embodiment of the present disclosure.

As shown in FIG. 8 , the method for screening an optimal strategy in this embodiment includes operation S810.

In operation S810, the n effective strategies are sorted according to the value of the first strategy effectiveness index in descending order, and the effective strategy with the first ranking is the optimal strategy.

According to another embodiment of the present disclosure, when there are q effective strategies that are ranked first, where q satisfies 2≤q≤n and q is an integer, the strategies of different effective strategies can also be updated by updating the traffic segmentation method Effectiveness index value to achieve optimal strategy screening.

FIG. 9 schematically shows a flowchart of a method for screening an optimal strategy according to another embodiment of the present disclosure.

As shown in FIG. 9 , the method for screening an optimal strategy according to another embodiment includes operation S910.

In operation S910, the traffic segmentation and updating method is performed i times until there is a first and only valid policy ranked first, and the valid policy is marked as the optimal policy, where i is an integer greater than or equal to 1.

FIG. 10 schematically shows a flowchart of a method for updating traffic segmentation according to an embodiment of the present disclosure.

As shown in FIG. 10 , the traffic segmentation and updating method in this embodiment includes operations S1010 to S1040.

In operation S1010, the traffic distribution ratio in the first traffic segmentation method is adjusted to obtain a second traffic segmentation method.

In operation S1020, user polling is allocated to the q effective policies based on the second traffic segmentation method.

In operation S1030, a second strategy effectiveness index value of the q types of effective strategies is obtained, and the second strategy effectiveness index value is calculated based on the same method as the first strategy effectiveness index value.

In operation S1040, the q effective strategies are sorted in descending order according to the second strategy effectiveness index value.

According to the embodiments of the present disclosure, the traffic distribution ratio in the first traffic segmentation method can be adjusted to obtain the second traffic segmentation method. For example, the first product recommendation strategy includes q kinds of product recommendation strategies. Assuming that q=4, the first product recommendation strategy is marked as Y ₀ , Y ₁ , Y ₂ , and Y ₃ respectively. In the first traffic segmentation method , the proportion of traffic allocated to each first product recommendation strategy is 25%. When obtaining the second traffic segmentation method, the traffic distribution ratio of each first product recommendation strategy can be adjusted to Y ₀ is 10%, Y ₁ is 30%, Y ₂ is 30%, and Y ₃ is 30%, thus Increased evaluation of dynamic bonus factors. After obtaining the second traffic segmentation method, user polling may be allocated to q effective strategies based on the second traffic segmentation method, and the second strategy effectiveness index may be calculated based on the same method as the first strategy effectiveness index value , and will not be repeated here. After sorting the q effective strategies according to the value of the second strategy effectiveness index from large to small, the optimal strategy can be obtained.

According to the embodiments of the present disclosure, after obtaining the optimal strategy, the optimal strategy can be continuously evaluated based on the full amount of user access data, and the operation results of the basic product recommendation strategy and the optimal strategy in a fixed time interval (for example, one month) can be compared , to re-verify the optimal strategy recommendation effect.

FIG. 11 schematically shows a flowchart of a method for evaluating an optimal strategy according to an embodiment of the present disclosure.

As shown in FIG. 11 , the method for evaluating the optimal strategy in this embodiment includes operations S1110 to S1130 .

In operation S1110, all users are allocated to the optimal strategy.

In operation S1120, an optimal policy effectiveness index value is calculated based on the same method as the first policy effectiveness index value.

In operation S1130, the optimal strategy is evaluated based on the optimal strategy effectiveness index value and a preset evaluation period.

In the embodiment of the present disclosure, by configuring multiple product recommendation strategies, the user traffic is segmented, and users are allocated to different product recommendation strategies according to the proportion of polling, so as to evaluate each product recommendation strategy, which is conducive to improving the The effectiveness of the product recommendation strategy, thereby increasing the transaction rate in the actual sales of the product. In this process, the optimal strategy can be obtained by updating the traffic segmentation, updating the relevant factors of the product recommendation strategy, etc., so as to facilitate the continuous operation and monitoring of the optimal strategy with all users, so as to continuously evaluate the effectiveness of the optimal strategy. This method can improve the penetration rate and success rate of online product marketing, recommend suitable online products to users when users have product purchase needs or regularly and timely, improve the convenience of customer transaction operations, improve customer stickiness, and help different customers to personalize products. Configuration planning.

Based on the above product recommendation method, the present disclosure also provides a product recommendation device. The device will be described in detail below with reference to FIG. 12 .

FIG. 12 schematically shows a structural block diagram of a product recommendation apparatus according to an embodiment of the present disclosure.

As shown in FIG. 12 , the product recommendation apparatus 1200 of this embodiment includes a first obtaining module 1210 , a first processing module 1220 , a second processing module 1230 , a second obtaining module 1240 and a third processing module 1250 .

The first obtaining module 1210 is configured to obtain a product recommendation strategy pool, where the product recommendation strategy pool includes m types of first product recommendation strategies, where m is an integer greater than or equal to 2.

The first processing module 1220 is configured to allocate user polling to the m first product recommendation strategies based on the product recommendation strategy pool and the first traffic segmentation method.

The second processing module 1230 is configured to generate a product recommendation list based on the first product recommendation policy assigned by the user and user access history data, wherein the user access history data includes the user's historical transaction product list and the user's historical browsing product list.

The second obtaining module 1240 is configured to obtain user access data associated with the product recommendation list.

The second processing module 1250 is configured to evaluate the m types of first product recommendation strategies based on a user access data set, wherein the user access data set includes a set of all user access data in the test period.

According to an embodiment of the present disclosure, any plurality of modules among the first obtaining module 1210, the first processing module 1220, the second processing module 1230, the second obtaining module 1240, and the third processing module 1250 may be combined in one module to implement, Or any one of the modules can be split into multiple modules. Alternatively, at least part of the functionality of one or more of these modules may be combined with at least part of the functionality of other modules and implemented in one module. According to an embodiment of the present disclosure, at least one of the first acquisition module 1210, the first processing module 1220, the second processing module 1230, the second acquisition module 1240, and the third processing module 1250 may be at least partially implemented as a hardware circuit, For example, Field Programmable Gate Array (FPGA), Programmable Logic Array (PLA), System on Chip, System on Substrate, System on Package, Application Specific Integrated Circuit (ASIC), or any other circuit that can be integrated or packaged by It can be implemented in hardware or firmware in a reasonable manner, or in any one of the three implementation manners of software, hardware and firmware, or in an appropriate combination of any of them. Alternatively, at least one of the first acquisition module 1210, the first processing module 1220, the second processing module 1230, the second acquisition module 1240, and the third processing module 1250 may be implemented at least partially as a computer program module, when the computer program When the module is run, it can perform the corresponding function.

FIG. 13 schematically shows a block diagram of an electronic device suitable for implementing a product recommendation method according to an embodiment of the present disclosure.

As shown in FIG. 13 , an electronic device 900 according to an embodiment of the present disclosure includes a processor 901 that can be loaded into a random access memory (RAM) 903 according to a program stored in a read only memory (ROM) 902 or from a storage portion 908 program to perform various appropriate actions and processes. The processor 901 may include, for example, a general-purpose microprocessor (eg, a CPU), an instruction set processor and/or a related chipset, and/or a special-purpose microprocessor (eg, an application-specific integrated circuit (ASIC)), and the like. The processor 901 may also include on-board memory for caching purposes. The processor 901 may include a single processing unit or multiple processing units for performing different actions of the method flow according to the embodiments of the present disclosure.

In the RAM 903, various programs and data necessary for the operation of the electronic device 900 are stored. The processor 901 , the ROM 902 and the RAM 903 are connected to each other through a bus 904 . The processor 901 performs various operations of the method flow according to the embodiment of the present disclosure by executing the programs in the ROM 902 and/or the RAM 903 . Note that the program may also be stored in one or more memories other than the ROM 902 and the RAM 903 . The processor 901 may also perform various operations of the method flow according to the embodiments of the present disclosure by executing programs stored in the one or more memories.

According to an embodiment of the present disclosure, the electronic device 900 may also include an input/output (I/O) interface 905 which is also connected to the bus 904 . Electronic device 900 may also include one or more of the following components connected to I/O interface 905: input portion 906 including keyboard, mouse, etc.; including components such as cathode ray tube (CRT), liquid crystal display (LCD), etc., and An output section 907 of speakers and the like; a storage section 908 including a hard disk and the like; and a communication section 909 including a network interface card such as a LAN card, a modem, and the like. The communication section 909 performs communication processing via a network such as the Internet. A drive 910 is also connected to the I/O interface 905 as needed. A removable medium 911, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is mounted on the drive 910 as needed so that a computer program read therefrom is installed into the storage section 908 as needed.

The present disclosure also provides a computer-readable storage medium. The computer-readable storage medium may be included in the device/apparatus/system described in the above embodiments; it may also exist alone without being assembled into the device/system. device/system. The above-mentioned computer-readable storage medium carries one or more programs, and when the above-mentioned one or more programs are executed, the product recommendation method according to the embodiment of the present disclosure is implemented.

According to an embodiment of the present disclosure, the computer-readable storage medium may be a non-volatile computer-readable storage medium, such as, but not limited to, portable computer disks, hard disks, random access memory (RAM), read only memory (ROM) , erasable programmable read only memory (EPROM or flash memory), portable compact disk read only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing. In this disclosure, a computer-readable storage medium may be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. For example, according to embodiments of the present disclosure, a computer-readable storage medium may include one or more memories other than ROM 902 and/or RAM 903 and/or ROM 902 and RAM 903 described above.

Embodiments of the present disclosure also include a computer program product comprising a computer program containing program code for performing the method illustrated in the flowchart. When the computer program product runs in the computer system, the program code is used to enable the computer system to implement the product recommendation method provided by the embodiments of the present disclosure.

When the computer program is executed by the processor 901, the above-described functions defined in the system/apparatus of the embodiment of the present disclosure are performed. According to embodiments of the present disclosure, the systems, apparatuses, modules, units, etc. described above may be implemented by computer program modules.

In one embodiment, the computer program may rely on a tangible storage medium such as an optical storage device, a magnetic storage device, or the like. In another embodiment, the computer program may also be transmitted, distributed in the form of a signal over a network medium, and downloaded and installed through the communication section 909, and/or installed from a removable medium 911. The program code embodied by the computer program may be transmitted using any suitable network medium, including but not limited to: wireless, wired, etc., or any suitable combination of the foregoing.

In such an embodiment, the computer program may be downloaded and installed from the network via the communication portion 909, and/or installed from the removable medium 911. When the computer program is executed by the processor 901, the above-described functions defined in the system of the embodiment of the present disclosure are performed. According to embodiments of the present disclosure, the above-described systems, apparatuses, apparatuses, modules, units, etc. can be implemented by computer program modules.

According to the embodiments of the present disclosure, the program code for executing the computer program provided by the embodiments of the present disclosure may be written in any combination of one or more programming languages, and specifically, high-level procedures and/or object-oriented programming may be used. programming language, and/or assembly/machine language to implement these computational programs. Programming languages include, but are not limited to, languages such as Java, C++, python, "C" or similar programming languages. The program code may execute entirely on the user computing device, partly on the user device, partly on a remote computing device, or entirely on the remote computing device or server. In the case of a remote computing device, the remote computing device may be connected to the user computing device through any kind of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (eg, using an Internet service provider business via an Internet connection).

The flowchart and block diagrams in the Figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code that contains one or more logical functions for implementing the specified functions executable instructions. It should also be noted that, in some alternative implementations, the functions noted in the blocks may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It is also noted that each block of the block diagrams or flowchart illustrations, and combinations of blocks in the block diagrams or flowchart illustrations, can be implemented in special purpose hardware-based systems that perform the specified functions or operations, or can be implemented using A combination of dedicated hardware and computer instructions is implemented.

Those skilled in the art will appreciate that various combinations and/or combinations of features recited in various embodiments and/or claims of the present disclosure are possible, even if such combinations or combinations are not expressly recited in the present disclosure. In particular, various combinations and/or combinations of the features recited in the various embodiments of the present disclosure and/or in the claims may be made without departing from the spirit and teachings of the present disclosure. All such combinations and/or combinations fall within the scope of this disclosure.

Embodiments of the present disclosure have been described above. However, these examples are for illustrative purposes only, and are not intended to limit the scope of the present disclosure. Although the various embodiments are described above separately, this does not mean that the measures in the various embodiments cannot be used in combination to advantage. The scope of the present disclosure is defined by the appended claims and their equivalents. Without departing from the scope of the present disclosure, those skilled in the art can make various substitutions and modifications, and these substitutions and modifications should all fall within the scope of the present disclosure.

Claims (20)

1. A method for recommending products, comprising:

obtaining a product recommendation strategy pool, wherein the product recommendation strategy pool comprises m first product recommendation strategies, and m is an integer greater than or equal to 2;

polling and distributing users to the m first product recommendation strategies based on the product recommendation strategy pool and a first traffic segmentation method;

generating a product recommendation list based on a first product recommendation strategy distributed by a user and user access historical data, wherein the user access historical data comprises a user historical transaction product list and a user historical browsing product list;

acquiring user access data, wherein the user access data is associated with a product recommendation list; and

evaluating the m first product recommendation strategies based on a user access data set, wherein the user access data set comprises a set of all user access data in a test period.

2. The method of claim 1, wherein the first product recommendation policy is generated based on a product base policy and a first dynamic addition factor.

3. The method of claim 2, wherein the product base policy comprises:

and calculating a basic product recommendation score for the to-be-recommended products contained in the to-be-recommended product list, wherein the basic product recommendation score is calculated based on the user transaction time factor, the user browsing time factor, the transaction product similarity and the browsing product similarity.

4. The method of claim 3, wherein the user transaction time factor and the user browsing time factor are calculated based on a time factor algorithm, the time factor algorithm comprising:

calculating a transaction time interval between a historical product transaction date and a current visit date, wherein the historical product transaction date is obtained based on the user historical transaction product list;

calculating the browsing time interval between the browsing date of the historical products and the current visiting date, wherein the browsing date of the historical products is obtained based on the historical browsing product list of the user;

acquiring the user transaction time factor based on the transaction time interval and the product deadline attribute; and

and acquiring the user browsing time factor based on the browsing time interval and the product term attribute.

5. The method of claim 3, wherein the method of generating the deal product similarity and the browse product similarity comprises:

calculating the similarity between the product to be recommended and the historical trading product of the user based on the product similarity index and the preset similarity index configuration, and acquiring the similarity of the trading product; and

and calculating the similarity between the product to be recommended and the historical browsing product of the user based on the product similarity index and the preset similarity index configuration, and acquiring the similarity of the browsing product.

6. The method of claim 5, wherein the product similarity index comprises two or more of product term, sales channel, risk level, purchase amount, product type, transaction currency, and credit status.

7. The method of claim 2, wherein the first dynamic addition factor comprises at least one trending factor, the trending factor comprising an asterisk trending factor, an age trending factor, or a risk assessment trending factor.

8. The method of claim 7, wherein the star popularity factor is determined based on historical access data of same-star users; the age trending factor is determined based on historical access data of users in the same age group; the risk evaluation popularity factor is determined based on historical access data of users with the same risk evaluation grade, wherein the user star grade, the user age group and the user risk evaluation grade are determined based on preset evaluation rules.

9. The method of claim 2, wherein the first product recommendation policy being generated based on a product base policy and a first dynamic addition factor comprises:

obtaining a first strategy weight vector, wherein the first strategy weight vector comprises a product basic strategy weight and a first dynamic addition factor weight; and

generating the first product recommendation policy based on the product base policy, the first dynamic addition factor, and the first policy weight vector.

10. The method of claim 1, wherein said evaluating the m first product recommendation policies based on a set of user access data further comprises:

calculating a strategy evaluation index value based on the user access data set, wherein the strategy evaluation index value comprises a product basic strategy evaluation index value and a first product recommendation strategy evaluation index value;

calculating a first policy validity index value based on the product base policy evaluation index value and the first product recommendation policy evaluation index value; and

when a first strategy effectiveness index value corresponding to the kth first product recommendation strategy is larger than or equal to a preset threshold value, marking the kth first product recommendation strategy as an effective strategy, wherein k is larger than or equal to 1 and is smaller than or equal to m, and k is an integer.

11. The method of claim 10, wherein when the first policy effectiveness index values corresponding to all of the first product recommendation policies are less than a preset threshold, the method further comprises:

executing the product recommendation strategy updating method for j times, marking the second product recommendation strategy as an effective strategy when a second strategy effectiveness index value corresponding to the second product recommendation strategy is larger than or equal to a preset threshold value, wherein j is an integer larger than or equal to 1,

the product recommendation strategy updating method comprises the following steps:

obtaining at least one of a second strategy weight vector or a second dynamic addition factor, wherein the second strategy weight vector is obtained by adjusting the relative proportion of the product basic strategy weight and the first dynamic addition factor weight in the first strategy weight vector, and the second dynamic addition factor is obtained by adjusting the type of the hot degree factor contained in the first dynamic addition factor;

generating a second product recommendation strategy based on the second strategy weight vector and/or a second dynamic addition factor and a product basic strategy; and

and evaluating the second product recommendation strategy based on the same evaluation method as the first product recommendation strategy.

12. The method of claim 10 or 11, wherein when n valid policies are included, where n satisfies 2 ≦ n ≦ m and n is an integer, the method further comprising:

and sequencing the n effective strategies from large to small according to the first strategy effectiveness index value, and taking the sequenced first effective strategy as an optimal strategy.

13. The method of claim 12, wherein when there are q active policies that are first ordered in parallel, where q satisfies 2 ≦ q ≦ n and q is an integer, the method further comprising:

executing the flow segmentation updating method for i times until a first and only effective strategy is sorted, marking the effective strategy as an optimal strategy, wherein i is an integer greater than or equal to 1,

the flow segmentation updating method comprises the following steps:

adjusting the flow distribution proportion in the first flow segmentation method to obtain a second flow segmentation method;

allocating user polling to the q effective strategies based on the second traffic segmentation method;

acquiring second strategy effectiveness index values of the q effective strategies, wherein the second strategy effectiveness index values are calculated based on the same method as the first strategy effectiveness index values;

and sequencing the q effective strategies from large to small according to a second strategy effectiveness index value.

14. The method of claim 12, wherein after obtaining the optimal policy, the method further comprises:

distributing a full amount of users to the optimal strategy;

calculating an optimal policy effectiveness index value based on the same method as the first policy effectiveness index value;

and evaluating the optimal strategy based on the optimal strategy effectiveness index value and a preset evaluation period.

15. The method of claim 10, wherein the user access data comprises page statistics and access behavior data, the page statistics including at least contact identification, product recommendation policy, policy invocation amount, and product recommendation number; the access behavior data is obtained based on at least one of a user transaction behavior and a user browsing behavior.

16. The method of claim 15, wherein the policy evaluation metrics include a recommended click rate and a purchase conversion rate.

17. A product recommendation device, comprising:

the system comprises a first obtaining module, a second obtaining module and a third obtaining module, wherein the first obtaining module is configured to obtain a product recommendation strategy pool, the product recommendation strategy pool comprises m first product recommendation strategies, and m is an integer greater than or equal to 2;

the first processing module is configured to distribute user polling to the m first product recommendation strategies based on the product recommendation strategy pool and a first traffic segmentation method;

the second processing module is configured to generate a product recommendation list based on the first product recommendation strategy distributed by the user and user access historical data, wherein the user access historical data comprises a user historical transaction product list and a user historical browsing product list;

the second acquisition module is configured to acquire user access data, and the user access data is associated with the product recommendation list; and

a third processing module configured to evaluate the m first product recommendation policies based on a user access data set, wherein the user access data set includes a set of all user access data in a test period.

18. An electronic device, comprising:

one or more processors;

a storage device for storing one or more programs,

wherein the one or more programs, when executed by the one or more processors, cause the one or more processors to perform the method of any of claims 1-16.

19. A computer readable storage medium having stored thereon executable instructions which, when executed by a processor, cause the processor to perform the method of any one of claims 1 to 16.

20. A computer program product comprising a computer program which, when executed by a processor, implements a method according to any one of claims 1 to 16.

Images