CN110599307B - A method and device for recommending products - Google Patents

Abstract

The application discloses a commodity recommending method and a commodity recommending device, in the method, a service platform can determine the attribute value expectation of a target commodity on a designated attribute through a pre-trained prediction model, the attribute value expectation indicates the attribute value of the target commodity expected by a user, in other words, the attribute value expectation effectively reflects the actual requirement of the user on the target commodity, so that the service platform can recommend the target commodity provided by a merchant conforming to the actual requirement of the user to the user based on the attribute value expectation.

Description

A method and device for recommending products

Technical Field

The embodiments of the present application relate to the field of computer technology, and more particularly to a method and device for recommending products.

Background Art

At present, the service platform can provide users with product recommendation services. While increasing the sales of various products on the service platform, it can also provide users with a variety of product choices, thus bringing users a good shopping experience.

In actual applications, the service platform can build a user profile of the user based on historical shopping records, browsing records, etc., and then use the user profile to recommend products in the product categories that the user likes and is interested in to the user in the subsequent process.

However, in the prior art, the service platform usually only selects some products from the product categories that the user is interested in and recommends them to the user. When the user searches for a specific product, the service platform usually cannot determine which merchant sells the specific product to recommend to the user according to the user's actual situation.

Therefore, how to select a merchant suitable for the user from among the many merchants selling the designated commodity based on the actual situation of the user, and recommend the designated commodity sold by the merchant to the user, is a problem to be solved urgently.

Summary of the invention

The embodiments of the present application provide a method and device for product recommendation to partially solve the above-mentioned problems existing in the prior art.

The present application embodiment adopts the following technical solutions:

The present application embodiment provides a method for recommending products, including:

Determining a target commodity to be recommended to the user based on the user identifier of the user;

Querying the attribute information corresponding to the target product, and querying the attribute information of the user according to the user identifier;

Inputting the attribute information of the user and the attribute information of the target product into a pre-trained prediction model to determine the probability distribution of the user placing an order for the target product whose attribute value of a specified attribute falls within different preset intervals;

Determining an expected attribute value of the target product on the specified attribute according to the probability distribution and the determined reference attribute values corresponding to the different preset intervals;

According to the commodity identifiers of the commodities provided by the merchants, query the merchants that provide the target commodity;

Determine the attribute value set by each queried merchant for the designated attribute of the target product;

For each attribute value, determining, according to the attribute value expectation, an order probability of the user placing an order for the target product having the attribute value;

According to each order probability determined for each attribute value, determine a merchant that needs to be recommended to the user from among the queried merchants as a target merchant;

The product information of the target product provided by the target merchant is determined as recommendation information, and based on the user identifier, the recommendation information is sent to a terminal held by the user for display, so as to recommend the target product provided by the target merchant to the user.

The present application embodiment provides a device for recommending products, including:

A commodity determination module, used to determine a target commodity to be recommended to the user based on the user identifier of the user;

A first query module, used to query the attribute information corresponding to the target product, and query the attribute information of the user according to the user identifier;

A probability distribution determination module, used to input the attribute information of the user and the attribute information of the target product into a pre-trained prediction model to determine the probability distribution of the user placing an order for the target product whose attribute value of a specified attribute falls within different preset intervals;

An expectation determination module, configured to determine an expectation of an attribute value of the target product on the specified attribute according to the probability distribution and the determined reference attribute values corresponding to the different preset intervals;

The second query module is used to query each merchant that provides the target commodity according to the commodity identifier of each commodity provided by each merchant;

An attribute value determination module, used to determine the attribute value set by each queried merchant for the specified attribute of the target product;

A probability determination module, configured to determine, for each attribute value, the probability of the user placing an order for the target product having the attribute value according to the attribute value expectation;

A merchant determination module, used to determine a merchant that needs to be recommended to the user as a target merchant from among the queried merchants according to the order placement probabilities determined for each attribute value;

The recommendation module is used to determine the commodity information of the target commodity provided by the target merchant as the recommendation information, and based on the user identifier, send the recommendation information to the terminal held by the user for display, so as to recommend the target commodity provided by the target merchant to the user.

An embodiment of the present application provides a computer-readable storage medium, wherein the storage medium stores a computer program, and when the computer program is executed by a processor, the above-mentioned method for recommending products is implemented.

An embodiment of the present application provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned product recommendation method when executing the program.

At least one of the above technical solutions adopted in the embodiments of the present application can achieve the following beneficial effects:

In the commodity recommendation method provided in the embodiment of the present application, based on the user's user identification, a target commodity recommended to the user is determined, attribute information corresponding to the target commodity is queried, and based on the user identification, attribute information of the user corresponding to the user identification is queried, and the queried attribute information corresponding to the target commodity and the attribute information of the user are input into a pre-trained prediction model to determine the probability distribution of the user placing an order for the target commodity whose attribute value of the specified attribute falls within different preset intervals, and based on the probability distribution and the reference attribute values corresponding to the determined different preset intervals, the attribute value expectation of the target commodity on the specified attribute is determined, and based on the commodities mentioned by each merchant, the attribute information of each merchant is input into a pre-trained prediction model to determine the probability distribution of the user placing an order for the target commodity whose attribute value of the specified attribute falls within different preset intervals. The product identification of the target product is used to query various merchants that provide the target product, determine the attribute values set by the queried merchants for the specified attributes of the target product, and for each attribute value, determine the order probability of the user placing an order for the target product with the attribute value according to the attribute value expectation, and according to each order probabilities determined for each attribute value, determine a merchant that needs to be recommended to the user from among the queried merchants as the target merchant, and then determine the product information of the target product provided by the target merchant as the recommendation information, and based on the user identification, send the recommendation information to the terminal held by the user for display, so as to recommend the target product provided by the target merchant to the user.

It can be seen from the above method that the service platform can determine the expected attribute value of the target product on the specified attribute through a pre-trained prediction model. The expected attribute value indicates the attribute value of the target product expected by the user. In other words, the expected attribute value effectively reflects the user's actual requirements for the target product. Therefore, the service platform subsequently recommends the target product provided by the merchant that meets the user's actual needs based on the expected attribute value. Compared with the existing technology, this can bring great convenience to the user.

BRIEF DESCRIPTION OF THE DRAWINGS

The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

FIG1 is a schematic diagram of a flow chart of a method for recommending products according to an embodiment of the present application;

FIG2 is a schematic diagram of recommending products to users through a determined Gaussian distribution according to an embodiment of the present application;

FIG3 is a schematic diagram of a device for recommending goods provided in an embodiment of the present application;

FIG. 4 is a schematic diagram of an electronic device corresponding to FIG. 1 provided in an embodiment of the present application.

DETAILED DESCRIPTION

In order to make the purpose, technical solution and advantages of the present application clearer, the technical solution of the present application will be clearly and completely described below in combination with the specific embodiments of the present application and the corresponding drawings. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present application.

The technical solutions provided by various embodiments of the present application are described in detail below in conjunction with the accompanying drawings.

FIG1 is a flow chart of a method for recommending products according to an embodiment of the present application, which specifically includes the following steps:

S101: Determine a target commodity to be recommended to the user based on the user identification of the user.

In order to enable users to get a better shopping experience, in an embodiment of the present application, the service platform can recommend products to users. Among them, there can be many specific triggering forms for the service platform to recommend products to users. For example, the user can enter the target information in a terminal device such as a mobile phone, a tablet computer, or an application (Application, App) installed in the terminal device. The terminal or App can receive the target information entered by the user and generate a corresponding information query request based on the user's user identifier (the user identifier can refer to identification information such as user account, mobile phone number, ID number, etc.). The terminal can send the information query request to the service platform, and accordingly, the service platform can determine the target information received by the terminal from the information query request carrying the user identifier, and determine the target product corresponding to the target information based on the target information. Among them, the target information mentioned here can refer to information used to identify the product, such as product name, product specifications, product category, etc. The target product mentioned here refers to the product that the user wants to query and needs to be recommended by the service platform.

For another example, when a user starts an App installed in a terminal, the terminal can send an information query request carrying a user identifier to the service platform corresponding to the App based on the user account logged in by the user in the App (the user identifier can refer to the user account, or can refer to identification information such as the user's mobile phone number, ID number, etc. queried through the user account). Correspondingly, the service platform can query the products that the user is interested in based on the user identifier carried in the received information query request, as well as the product categories and products that each user is interested in that have been pre-counted, and then these queried products can be determined as target products that need to be recommended to the user.

For another example, the service platform can actively recommend products to the user, that is, the user does not need to send an information query request to the service platform through a terminal or App. The service platform can actively query the products that the user is interested in based on the user's user identification (i.e. user account, mobile phone number, ID number and other identification information) and the pre-counted product categories and products that each user is interested in, and then determine these queried products as target products that need to be recommended to the user. Other trigger forms are not given examples here one by one.

It should be noted that when determining the target products that need to be recommended to the user based on the pre-counted product categories and products that each user is interested in, it may eventually be determined that multiple target products need to be recommended to the user. For each target product, the service platform can recommend the product to the user through the product recommendation method provided in the embodiment of the present application.

S102: Query the attribute information corresponding to the target product, and query the attribute information of the user according to the user identifier.

S103: Inputting the attribute information of the user and the attribute information of the target product into a pre-trained prediction model to determine the probability distribution of the user placing an order for the target product whose attribute value of a specified attribute falls within different preset intervals.

After determining the target product corresponding to the target information, the service platform can query the attribute information corresponding to the target product based on the product identifier of the target product, and query the attribute information of the user corresponding to the user identifier based on the user identifier carried in the information query request. Among them, the user's attribute information may refer to a user portrait, and the attribute information may reflect user characteristics such as user preferences and the average amount of money spent by the user on each type of product. Similarly, the attribute information of the target product mentioned here may refer to the product portrait of the target product, and the attribute information of the target product may effectively reflect characteristics such as the product type, product sales volume, and product price of the target product.

The service platform inputs the attribute information of the user and the attribute information of the target product into a pre-trained prediction model, thereby determining the probability distribution of the user placing an order for the target product whose attribute value of the specified attribute falls within different preset intervals. Among them, the specified attribute may include product attributes such as price and specifications of the product. For ease of description, the method for product recommendation provided in the embodiment of the present application will be described below only with the price of the product as the specified attribute. Accordingly, when the specified attribute is the price of the product, the attribute value of the specified attribute refers to the specific selling price of the product.

The above probability distribution is mainly used to characterize the probability of the user placing an order for the target product in different price ranges. In the embodiment of the present application, the specific selling prices of the target product sold by each merchant can be counted in advance, and the specific selling prices can be sorted from low to high by artificial division to obtain a price sequence, and then the price sequence is divided into a preset number of price intervals to obtain each preset interval. Of course, the highest price and the lowest price of the target product sold by each merchant can also be counted in advance, and the price range from the lowest price to the highest price can be divided into a preset number of price intervals to obtain each preset interval.

Through the above prediction model, the service platform can determine the probability distribution of the user placing an order for the target product in each preset interval. For example, the probability distribution determined by the service platform through the prediction model is: [0.08, 0.15, 0.26, 0.48, 0.03], where 0.08 represents the probability of the user placing an order for product A (i.e., the target product) whose specific price falls within the first preset interval, 0.15 represents the probability of the user placing an order for product A (i.e., the target product) whose specific price falls within the second preset interval, and so on.

S104: Determine an expected attribute value of the target product on the specified attribute according to the probability distribution and the determined reference attribute values corresponding to the different preset intervals.

The service platform can determine the expected attribute value of the target product on the specified attribute based on the determined probability distribution. Specifically, the service platform can determine the reference attribute values corresponding to different preset intervals. For each preset interval, the reference attribute value corresponding to the preset interval is used to characterize the size of the attribute value in the preset interval as a whole.

In an embodiment of the present application, the service platform can determine the reference attribute value by determining an average value. Specifically, for each preset interval, the service platform can determine the average value of each attribute value of the commodity within the preset interval as the reference attribute value corresponding to the preset interval. For example, assuming the preset interval is: [19.0, 19.5, 19.8, 20.5, 21.3], the service platform can average the specific selling prices contained in the preset interval, thereby obtaining an average value of: 20.02, and then 20.02 can be used as the reference attribute value corresponding to the preset interval.

The service platform can also determine the reference attribute value by determining the median. Specifically, for each preset interval, the service platform can determine the median of each attribute value of the target product within the preset interval as the reference attribute value corresponding to the preset interval. Continuing with the above example, in the preset interval [19.0, 19.5, 19.8, 20.5, 21.3], the median is 19.8, then the service platform can use the median 19.8 as the reference attribute value corresponding to the preset interval.

The service platform can determine the reference attribute value corresponding to each preset interval in the manner described above, and then determine the expected attribute value of the target product on the specified attribute based on the above probability distribution and the reference attribute value corresponding to each preset interval. Specifically, for each preset interval, the service platform can determine the probability corresponding to the preset interval based on the probability distribution, and then use the product of the probability corresponding to the preset interval and the reference attribute value corresponding to the preset interval as the first product corresponding to the preset interval. The service platform can add up the first products corresponding to each preset interval to determine the expected attribute value of the target product on the specified attribute. Among them, the product of the probability corresponding to the preset interval and the reference attribute value corresponding to the preset interval is called the first product, which is mainly used to distinguish it from the second product mentioned later, and there is no other special meaning for "first" and "second" themselves.

For example, suppose that five preset intervals are pre-divided for commodity B, and the reference attribute values corresponding to these five preset intervals are: 19.8, 21.3, 22.2, 23.1, and 24.1. The service platform determines through the prediction model that the probability distribution of the user placing an order for commodities with specific prices falling within different preset intervals is: [0.08, 0.15, 0.26, 0.48, 0.03]. The service platform can determine the expected attribute value through the following formula.

Wherein, P _k represents the probability corresponding to the kth preset interval in the probability distribution, Price _k represents the reference attribute value corresponding to the kth preset interval, and Exp represents the determined attribute value expectation.

The attribute value expectation determined by the service platform is used to characterize the attribute value expected by the user when purchasing the target product. That is, when the specified attribute is the price of the product, the attribute value expectation actually refers to the specific selling price expected by the user when purchasing the target product. The service platform can subsequently determine the merchant whose specific selling price for selling the target product is close to the attribute value expectation from the merchants on the service platform through the attribute value expectation, and recommend the target product sold by the merchant to the user.

S105: Querying merchants that provide the target commodity based on the commodity identifiers of the commodities provided by the merchants.

S106: Determine the attribute value set by each queried merchant for the designated attribute of the target product.

In this specification, the service platform records the commodities provided by each merchant, that is, records which commodities each merchant sells. Therefore, the service platform can query the merchants that provide the target commodity according to the commodity identifiers of the recorded commodities provided by each merchant (that is, the commodities sold by each merchant). In other words, the service platform can query which merchants sell the target commodity.

Furthermore, the service platform can determine the attribute values set by each merchant for the designated attribute of the target product according to the recorded attribute values set by each merchant for the products they provide. That is, when the designated attribute is the price of the product, the service platform needs to determine the specific selling price (i.e., attribute value) of each merchant selling the target product for the target product.

It should be noted that the service platform can perform step S104 at many times. The service platform can perform step S104 after determining the target product corresponding to the target information, after determining the attribute value expectation, or after determining the above-mentioned probability distribution. Other times are not described one by one here.

S107: For each attribute value, according to the attribute value expectation, determining the order probability of the user placing an order for the target product having the attribute value.

After the service platform determines the attribute values of the designated attributes of the target product provided by each merchant, it can compare the attribute value with the determined attribute value expectation for each attribute value, and then determine the probability of the user placing an order for the target product with the attribute value. Among them, if the attribute value is closer to the attribute value expectation in terms of value, the probability of the user placing an order for the target product with the attribute value is higher, and if the attribute value is more different from the attribute value expectation in terms of value, the probability of the user placing an order for the target product with the attribute value is lower.

In an embodiment of the present application, for each attribute value, the service platform can determine the probability of a user placing an order for the target product having the attribute value through a preset algorithm. Specifically, the service platform can use the attribute value and the attribute value expectation as the input of the preset algorithm, and the output of the preset algorithm is the probability of a user placing an order for the target product having the attribute value. It should be pointed out that the specific form of the preset algorithm can be many, and no specific limitation is made here. It only needs to ensure that the size of the difference between the attribute value and the attribute value expectation is negatively correlated with the size of the determined order probability.

S108: According to each ordering probability determined for each attribute value, determine a merchant that needs to be recommended to the user from among the queried merchants as a target merchant.

S109: Determine the product information of the target product provided by the target merchant as recommendation information, and send the recommendation information to the terminal held by the user for display based on the user identifier, so as to recommend the target product provided by the target merchant to the user.

The service platform can sort the merchants that provide the target product from high to low according to the determined order probabilities, and use the merchants ranked before the set ranking as target merchants, and then recommend the target merchant and the target product provided by the target merchant to the user. Among them, the service platform can send the product information of the target product provided by the target merchant as recommendation information to the terminal held by the user (such as the terminal of the currently logged-in user account), so that the terminal displays the recommendation information, or displays the recommendation information through the corresponding App. In this way, the service platform recommends the target product provided by the target merchant to the user.

Of course, after determining the order probabilities, the service platform can also determine the target merchant in other ways. For example, the service platform can score each merchant according to the determined order probabilities, the number of views of each merchant, the sales volume of each merchant, the star rating of each merchant, and the preset weights, and then use the merchants with higher scores as target merchants, and recommend the target merchants and the target products provided by the target merchants to the user; for another example, the service platform can first sort the merchants in order of the order probabilities from high to low, and then the service platform can select the merchants with complaints below the set threshold from the merchants ranked before the set ranking as the target merchants, and then recommend the target merchants and the target products provided by the target merchants to the user. Other methods will not be explained in detail here.

From the above method, it can be seen that the service platform can determine the expected attribute value of the target product on the specified attribute through the pre-trained prediction model. The expected attribute value indicates the attribute value of the target product expected by the user. In other words, the expected attribute value effectively reflects the actual requirements of the user for the target product. Therefore, the service platform subsequently recommends the target product provided by the merchant that meets the actual needs of the user based on the expected attribute value, thereby bringing a good shopping experience and great convenience to the user.

In the embodiment of the present application, after the service platform determines the above-mentioned attribute value expectation, it can further determine the attribute value variance of the target product on the specified attribute based on the attribute value expectation, and then determine the Gaussian distribution of the target product on the specified attribute based on the attribute value expectation and the attribute value variance. The service platform can determine the attribute value variance of the target product on the specified attribute based on the determined probability distribution, the reference attribute values corresponding to different preset intervals, and the attribute value expectation.

Specifically, for each preset interval, the service platform can determine the difference between the reference attribute value corresponding to the preset interval and the expected attribute value, and then determine the product of the probability corresponding to the preset interval on the probability distribution and the square of the difference, and use the product as the second product corresponding to the preset interval. The service platform can determine the sum of the second products corresponding to different preset intervals as the attribute value variance. Taking the price of the commodity as an example, the following formula can be used for specific reference:

In this formula, P _k represents the probability corresponding to the kth preset interval in the probability distribution, Price _k represents the reference attribute value corresponding to the kth preset interval, Exp represents the determined attribute value expectation, and Var ² represents the determined attribute value variance.

After determining the variance of the attribute value, the corresponding Gaussian distribution can be constructed: N(Exp, Var ² ). Then, in the subsequent process, the Gaussian distribution is used to determine the order probabilities and recommend products to the user, as shown in FIG2 .

FIG2 is a schematic diagram of recommending products to users through a determined Gaussian distribution according to an embodiment of the present application.

Assume that the service platform finds out that merchant A sets the price of product a (i.e., the target product) as 30. The service platform can determine the probability of a user placing an order for product a when the price of product a is 30 from the determined Gaussian distribution. Similarly, the service platform determines the order probabilities of users placing orders for product a at different prices in this way, and determines the merchant that needs to be recommended to the user and the product a sold by the merchant through each order probability.

In the embodiment of the present application, the attribute value expectation determined by the service platform is used to characterize the attribute value of the target product that the user expects when purchasing the target product, so the attribute value expectation is actually corresponding to the user and the target product, and the attribute value variance is determined by the attribute value expectation, so the attribute value variance is also corresponding to the user and the target product. Further, the Gaussian distribution determined by the attribute value variance and the attribute value expectation is also corresponding to the user and the target product.

In other words, the Gaussian distribution determined by the service platform is related to the user and the product that the user wants to buy, and the Gaussian distribution is consistent with the actual situation of the user and the target product. Based on this, the service platform can recommend products that meet the actual needs of the user through the Gaussian distribution, thereby bringing great convenience and good user experience to the user.

In the embodiment of the present application, the service platform can pre-train the above prediction model. Specifically, the service platform can obtain the historical order data of each user as a training sample, and use the preset intervals in which the attribute values of the specified attributes of the goods contained in these historical order data fall as training targets to train the prediction model. Since the training method of the prediction model is an existing conventional method, the training process of the prediction model is not described in detail here.

It should also be noted that the above is an example of the price of a commodity with a specified attribute, and the method for recommending commodities provided by the present application is described. The commodity recommendation method provided by the embodiment of the present application is also applicable to the case where the specified attribute is the attribute of other commodities. For example, when the target commodity is a storage box and the specified attribute is the storage space size of the storage box, the above preset interval is pre-divided according to the storage space size of each storage box sold by each merchant. The service platform can determine the probability distribution of the user placing an order for a storage box with a storage space size within different preset intervals through a pre-trained prediction model, and according to the probability distribution, determine the storage space expectation (i.e., attribute value expectation) of the user for the storage box, and then recommend the merchant whose storage space size of the storage box sold is close to the storage space expectation to the user, and recommend the storage box sold by the merchant to the user. The specific process is basically the same as above, and will not be described in detail here.

The above is a method for recommending goods provided by one or more embodiments of the present application. Based on the same idea, the embodiments of the present application also provide a corresponding device for recommending goods, as shown in FIG. 3 .

FIG3 is a schematic diagram of a device for recommending products provided in an embodiment of the present application, which specifically includes:

A commodity determination module 301 is used to determine a target commodity recommended to the user based on the user identifier of the user;

The first query module 302 is used to query the attribute information corresponding to the target product, and query the attribute information of the user corresponding to the user identifier according to the user identifier carried in the information query request;

The probability distribution determination module 303 is used to input the attribute information of the user and the attribute information of the target product into a pre-trained prediction model to determine the probability distribution of the user placing an order for the target product whose attribute value of the specified attribute falls within different preset intervals;

An expectation determination module 304 is used to determine the attribute value expectation of the target product on the specified attribute according to the probability distribution and the determined reference attribute values corresponding to the different preset intervals;

The second query module 305 is used to query each merchant that provides the target commodity according to the commodity identifier of each commodity provided by each merchant;

The attribute value determination module 306 is used to determine the attribute value set by each queried merchant for the designated attribute of the target product;

A probability determination module 307, configured to determine, for each attribute value, according to the attribute value expectation, the probability of the user placing an order for the target product having the attribute value;

A merchant determination module 308, configured to determine a merchant that needs to be recommended to the user as a target merchant from among the merchants found according to the order placement probabilities determined for each attribute value;

The recommendation module 309 is used to determine the product information of the target product provided by the target merchant as recommendation information, and send the recommendation information to the terminal held by the user for display based on the user identifier to recommend the target product provided by the target merchant to the user.

Optionally, the expectation determination module 304 is specifically used to, for each preset interval, determine the average value of each attribute value of the target product within the preset interval as the reference attribute value corresponding to the preset interval; or for each preset interval, determine the median of each attribute value of the target product within the preset interval as the reference attribute value corresponding to the preset interval.

Optionally, the expectation determination module 304 is specifically used to, for each preset interval, determine the probability corresponding to the preset interval according to the probability distribution; take the product of the probability corresponding to the preset interval and the reference attribute value corresponding to the preset interval as the first product corresponding to the preset interval; and determine the sum of the first products corresponding to the preset intervals as the attribute value expectation.

Optionally, the probability determination module 307 is specifically used to determine the attribute value variance of the target product on the specified attribute based on the attribute value expectation; determine the Gaussian distribution of the target product on the specified attribute based on the attribute value expectation and the attribute value variance; for each attribute value, determine the order probability of the user placing an order for the target product with the attribute value based on the Gaussian distribution.

Optionally, the probability determination module 307 is specifically configured to determine the attribute value variance of the target product on the specified attribute according to the probability distribution, reference attribute values corresponding to different preset intervals, and the attribute value expectation.

Optionally, the probability determination module 307 is specifically used to determine, for each preset interval, the difference between the reference attribute value corresponding to the preset interval and the expected attribute value; determine the product of the probability corresponding to the preset interval on the probability distribution and the square of the difference, and use the product as the second product corresponding to the preset interval; and determine the sum of the second products corresponding to different preset intervals as the attribute value variance.

Optionally, the merchant determination module 308 is specifically configured to sort the queried merchants from high to low according to the determined order placement probabilities from large to small, and select merchants ranked before a set ranking as target merchants.

Optionally, the specified attributes include: price of the product.

An embodiment of the present application also provides a computer-readable storage medium, which stores a computer program. The computer program can be used to execute the product recommendation method provided in FIG. 1 above.

The embodiment of the present application also provides a schematic structural diagram of an electronic device shown in FIG4. As shown in FIG4, at the hardware level, the electronic device includes a processor, an internal bus, a network interface, a memory, and a non-volatile memory, and of course may also include hardware required for other services. The processor reads the corresponding computer program from the non-volatile memory into the memory and then runs it to implement the method for recommending goods described in FIG1 above. Of course, in addition to software implementations, this specification does not exclude other implementations, such as logic devices or a combination of software and hardware, etc., that is to say, the execution subject of the following processing flow is not limited to each logic unit, but can also be hardware or logic devices.

In the 1990s, improvements to a technology could be clearly distinguished as hardware improvements (for example, improvements to the circuit structure of diodes, transistors, switches, etc.) or software improvements (improvements to the method flow). However, with the development of technology, many improvements to the method flow today can be regarded as direct improvements to the hardware circuit structure. Designers almost always obtain the corresponding hardware circuit structure by programming the improved method flow into the hardware circuit. Therefore, it cannot be said that an improvement in a method flow cannot be implemented using a hardware entity module. For example, a programmable logic device (PLD) (such as a field programmable gate array (FPGA)) is such an integrated circuit whose logical function is determined by the user's programming of the device. Designers can "integrate" a digital system on a PLD by programming it themselves, without having to ask a chip manufacturer to design and produce a dedicated integrated circuit chip. Moreover, nowadays, instead of manually making integrated circuit chips, this kind of programming is mostly implemented by "logic compiler" software, which is similar to the software compiler used when developing and writing programs, and the original code before compilation must also be written in a specific programming language, which is called hardware description language (HDL). There is not only one HDL, but many kinds, such as ABEL (Advanced Boolean Expression Language), AHDL (Altera Hardware Description Language), Confluence, CUPL (Cornell University Programming Language), HDCal, JHDL (Java Hardware Description Language), Lava, Lola, MyHDL, PALASM, RHDL (Ruby Hardware Description Language), etc. The most commonly used ones are VHDL (Very-High-Speed Integrated Circuit Hardware Description Language) and Verilog. Those skilled in the art should also know that it is only necessary to program the method flow slightly in the above-mentioned hardware description languages and program it into the integrated circuit, and then it is easy to obtain the hardware circuit that implements the logic method flow.

The controller can be implemented in any appropriate manner, for example, the controller can take the form of a microprocessor or processor and a computer-readable medium storing a computer-readable program code (such as software or firmware) that can be executed by the (micro)processor, a logic gate, a switch, an application-specific integrated circuit (ASIC), a programmable logic controller, and an embedded microcontroller. Examples of controllers include, but are not limited to, the following microcontrollers: ARC 625D, Atmel AT91SAM, Microchip PIC18F26K20, and Silicone Labs C8051F320. The memory controller can also be implemented as part of the control logic of the memory. Those skilled in the art also know that in addition to implementing the controller in a purely computer-readable program code manner, the controller can be implemented in the form of a logic gate, a switch, an application-specific integrated circuit, a programmable logic controller, and an embedded microcontroller by logically programming the method steps. Therefore, this controller can be considered as a hardware component, and the devices included therein for implementing various functions can also be regarded as structures within the hardware component. Or even, the devices for implementing various functions can be regarded as both software modules for implementing the method and structures within the hardware component.

The systems, devices, modules or units described in the above embodiments may be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, the computer may be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

For the convenience of description, the above device is described in various units according to their functions. Of course, when implementing the present application, the functions of each unit can be implemented in the same or multiple software and/or hardware.

Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

The present invention is described with reference to the flowchart and/or block diagram of the method, device (system), and computer program product according to the embodiment of the present invention. It should be understood that each process and/or box in the flowchart and/or block diagram, as well as the combination of the process and/or box in the flowchart and/or block diagram can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device produce a device for implementing the functions specified in one or more processes in the flowchart and/or one or more boxes in the block diagram.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured product including an instruction device that implements the functions specified in one or more processes in the flowchart and/or one or more boxes in the block diagram.

These computer program instructions may also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, whereby the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one or more processes in the flowchart and/or one or more boxes in the block diagram.

In a typical configuration, a computing device includes one or more processors (CPU), input/output interfaces, network interfaces, and memory.

The memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and/or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

The present application may be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The present application may also be practiced in distributed computing environments where tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules may be located in local and remote computer storage media, including storage devices.

Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

The above is only an embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.

Claims (11)

1. A method of merchandise recommendation, comprising:

determining a target commodity recommended to a user based on a user identification of the user;

inquiring attribute information corresponding to the target commodity, and inquiring the attribute information of the user according to the user identification;

Inputting the attribute information of the user and the attribute information of the target commodity into a pre-trained prediction model to determine probability distribution of the user for ordering the target commodity with the attribute value of the designated attribute falling into different preset intervals;

determining an attribute value expectation of the target commodity on the designated attribute according to the probability distribution and the determined reference attribute values corresponding to the different preset intervals respectively;

Inquiring each merchant providing the target commodity according to the commodity identification of each commodity provided by each merchant;

determining attribute values set by each queried merchant for the designated attributes of the target commodity;

For each attribute value, determining the ordering probability of the user to order the target commodity with the attribute value according to the attribute value expectation;

Determining a merchant needing to be recommended to the user from queried merchants according to each ordering probability determined for each attribute value, and taking the merchant as a target merchant;

And determining commodity information of the target commodity provided by the target merchant as recommendation information, and sending the recommendation information to a terminal held by the user for display based on the user identification so as to recommend the target commodity provided by the target merchant to the user.

2. The method of claim 1, wherein determining the reference attribute values corresponding to the different preset intervals respectively specifically includes:

For each preset interval, determining the average value of the attribute values of the target commodity in the preset interval as a reference attribute value corresponding to the preset interval; or (b)

And aiming at each preset interval, determining the median of each attribute value of the target commodity in the preset interval as a reference attribute value corresponding to the preset interval.

3. The method of claim 1, wherein determining the expected attribute value of the target commodity on the specified attribute according to the probability distribution and the determined reference attribute values respectively corresponding to the different preset intervals, specifically comprises:

for each preset interval, determining the probability corresponding to the preset interval according to the probability distribution;

Taking the product of the probability corresponding to the preset interval and the reference attribute value corresponding to the preset interval as a first product corresponding to the preset interval;

and determining the sum value of the first products corresponding to the preset intervals as the expected attribute value.

4. A method according to claim 3, wherein for each attribute value, determining the ordering probability of the user for the target commodity having that attribute value based on the attribute value expectations, specifically comprises:

Determining the attribute value variance of the target commodity on the appointed attribute according to the attribute value expectation;

Determining Gaussian distribution of the target commodity on the appointed attribute according to the attribute value expectation and the attribute value variance;

And determining the ordering probability of the user for ordering the target commodity with the attribute value according to the Gaussian distribution for each attribute value.

5. The method of claim 4, wherein determining the variance of the attribute value for the target commodity over the specified attribute based on the attribute value expectation, comprises:

And determining the attribute value variance of the target commodity on the designated attribute according to the probability distribution, the reference attribute values corresponding to different preset intervals and the attribute value expectations.

6. The method of claim 5, wherein determining the variance of the attribute value of the target commodity on the specified attribute according to the probability distribution, the reference attribute values corresponding to different preset intervals, and the attribute value expectations, comprises:

for each preset interval, determining a difference value between a reference attribute value corresponding to the preset interval and the attribute value expectation;

Determining the product of the probability corresponding to the preset interval on the probability distribution and the square of the difference value, and taking the product as a second product corresponding to the preset interval;

And determining the sum of the second products corresponding to different preset intervals as the attribute value variance.

7. The method of claim 1, wherein determining, as target merchants, merchants that need to be recommended to the user from among the queried merchants according to the respective ordering probabilities determined for each attribute value, specifically comprises:

and sequencing the queried merchants from high to low according to the determined order of the probability of each order from high to low, and taking the merchants ranked before the ranking setting as target merchants.

8. The method of any of claims 1-7, wherein the specifying the attribute comprises: price of the commodity.

9. An apparatus for recommending goods, comprising:

The commodity determining module is used for determining target commodities recommended to a user based on the user identification of the user;

The first query module is used for querying attribute information corresponding to the target commodity and querying the attribute information of the user according to the user identifier;

The probability distribution determining module is used for inputting the attribute information of the user and the attribute information of the target commodity into a pre-trained prediction model so as to determine probability distribution of the user for ordering the target commodity with the attribute value of the designated attribute falling into different preset intervals;

the expected determining module is used for determining the expected attribute value of the target commodity on the appointed attribute according to the probability distribution and the determined reference attribute values corresponding to the different preset intervals respectively;

the second inquiry module is used for inquiring each merchant providing the target commodity according to the commodity identification of each commodity provided by each merchant;

The attribute value determining module is used for determining the attribute value set by each queried merchant for the designated attribute of the target commodity;

The probability determining module is used for determining the ordering probability of the user for ordering the target commodity with the attribute value according to the attribute value expectation;

the merchant determining module is used for determining merchants needing to be recommended to the user from queried merchants according to the probability of each ordering determined for each attribute value, and taking the merchants as target merchants;

And the recommending module is used for determining commodity information of the target commodity provided by the target merchant, sending the recommending information to a terminal held by the user for displaying based on the user identification so as to recommend the target commodity provided by the target merchant to the user.

10. A computer-readable storage medium, characterized in that the storage medium stores a computer program which, when executed by a processor, implements the method of any of the preceding claims 1-8.

11. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the method of any of the preceding claims 1-8 when executing the program.