CN110675217A - Personalized background image generation method and device - Google Patents
Personalized background image generation method and device Download PDFInfo
- Publication number
- CN110675217A CN110675217A CN201910836136.9A CN201910836136A CN110675217A CN 110675217 A CN110675217 A CN 110675217A CN 201910836136 A CN201910836136 A CN 201910836136A CN 110675217 A CN110675217 A CN 110675217A
- Authority
- CN
- China
- Prior art keywords
- matrix
- user
- commodity
- access
- feature
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 77
- 239000011159 matrix material Substances 0.000 claims abstract description 213
- 239000013598 vector Substances 0.000 claims abstract description 115
- 238000004364 calculation method Methods 0.000 claims description 17
- 238000013507 mapping Methods 0.000 claims description 12
- 238000013480 data collection Methods 0.000 claims description 4
- 238000000605 extraction Methods 0.000 claims description 4
- 238000012545 processing Methods 0.000 description 15
- 238000004590 computer program Methods 0.000 description 10
- 239000000284 extract Substances 0.000 description 9
- 238000010586 diagram Methods 0.000 description 6
- 230000000694 effects Effects 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 4
- 238000010606 normalization Methods 0.000 description 4
- 230000001360 synchronised effect Effects 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000009877 rendering Methods 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0641—Shopping interfaces
- G06Q30/0643—Graphical representation of items or shoppers
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0241—Advertisements
- G06Q30/0251—Targeted advertisements
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0631—Item recommendations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0633—Lists, e.g. purchase orders, compilation or processing
- G06Q30/0635—Processing of requisition or of purchase orders
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Theoretical Computer Science (AREA)
- Strategic Management (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Development Economics (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Economics (AREA)
- Data Mining & Analysis (AREA)
- Entrepreneurship & Innovation (AREA)
- Game Theory and Decision Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Information Transfer Between Computers (AREA)
Abstract
The application relates to a personalized background image generation method and device. The method comprises the following steps: collecting access data and order data of a plurality of users; establishing a first characteristic matrix according to the access data of each user, and establishing a second characteristic matrix according to the order data of each user; extracting a first feature vector from the first feature matrix, and extracting a second feature vector from the second feature matrix; matching in the basic user group matrix according to the first eigenvector to obtain a first classification group, and matching in the basic user group matrix according to the second eigenvector to obtain a second classification group; and obtaining corresponding recommended commodities and generating background posters according to the first classification group and the second classification group. By adopting the method, the accuracy of recommending commodities can be improved, and the advertisement promotion efficiency can be improved.
Description
Technical Field
The present application relates to the field of internet data processing technologies, and in particular, to a method and an apparatus for generating a personalized background map.
Background
Internet shopping is a shopping mode in which a shopping web page or shopping software is registered through a terminal device, and after browsing and selecting goods to be purchased, an electronic shopping request is issued and a shopping order is generated. With the popularity of the internet and the convenience of internet shopping, more and more users prefer to select internet shopping as their primary shopping mode. In the internet shopping process, a merchant webpage or shopping software often recommends a commodity which is of interest to a user in a page pop-up advertisement mode so as to improve the conversion rate of advertisement pushing.
In the conventional technology, a merchant webpage or shopping software generally uses a part of commodities with the best sales volume as commodities interested by a user, or uses historical order data of the user as a basis, uses commodities of the same type as the historical order data as the commodities interested by the user, and pushes the commodities to the user in a pop-up page mode.
However, in practical cases, other products of the same type or products with the best sales volume are recommended according to the products purchased by the user, and since some types of products do not purchase other products of the same type again after purchase, the products with the best sales volume cannot reflect the preference of the current user. Therefore, the recommendation method in the traditional technology cannot meet the actual interest needs of the users, so that the advertisement promotion efficiency is low, and the mode of popping up the page can also make the users bored and the promotion effect is poor.
Disclosure of Invention
Therefore, it is necessary to provide a method, an apparatus, a computer device and a storage medium for generating a personalized background map, which can accurately reflect the user's preference, thereby improving the advertisement promotion efficiency and the effect.
In a first aspect, the present application provides a method for generating a personalized background map, where the method includes:
collecting historical data of a plurality of users; the historical data comprises access data and order data;
establishing a first characteristic matrix according to the access data of each user, and establishing a second characteristic matrix according to the order data of each user;
extracting a first feature vector of a target user from the first feature matrix, and extracting a second feature vector of the target user from the second feature matrix;
matching in the basic user group matrix according to the first eigenvector to obtain a first classification group, and matching in the basic user group matrix according to the second eigenvector to obtain a second classification group;
acquiring corresponding recommended commodities according to the first classification group and the second classification group;
and generating a background poster according to the recommended commodity.
In a second aspect, the present application provides a personalized background map generation apparatus, including:
the data collection module is used for collecting historical data of a plurality of users; the historical data comprises access data and order data;
the characteristic matrix establishing module is used for establishing a first characteristic matrix according to the access data of each user and establishing a second characteristic matrix according to the order data of each user;
the feature vector extraction module is used for extracting a first feature vector of a target user from the first feature matrix and extracting a second feature vector of the target user from the second feature matrix;
the matching module is used for matching in a basic user group matrix according to the first characteristic vector to obtain a first classification group and matching in the basic user group matrix according to the second characteristic vector to obtain a second classification group;
the recommending module is used for acquiring corresponding recommended commodities according to the first classification group and the second classification group;
and the poster generating module is used for generating a background poster according to the recommended commodity.
In a third aspect, the present application provides a computer device comprising a memory and a processor, the memory storing a computer program, the processor implementing the following steps when executing the computer program:
collecting historical data of a plurality of users; the historical data comprises access data and order data; establishing a first characteristic matrix according to the access data of each user, and establishing a second characteristic matrix according to the order data of each user; extracting a first feature vector of a target user from the first feature matrix, and extracting a second feature vector of the target user from the second feature matrix; matching in the basic user group matrix according to the first eigenvector to obtain a first classification group, and matching in the basic user group matrix according to the second eigenvector to obtain a second classification group; acquiring corresponding recommended commodities according to the first classification group and the second classification group; and generating a background poster according to the recommended commodity.
In a fourth aspect, the present application provides a computer readable storage medium having stored thereon a computer program which, when executed by a processor, performs the steps of:
collecting historical data of a plurality of users; the historical data comprises access data and order data; establishing a first characteristic matrix according to the access data of each user, and establishing a second characteristic matrix according to the order data of each user; extracting a first feature vector of a target user from the first feature matrix, and extracting a second feature vector of the target user from the second feature matrix; matching in the basic user group matrix according to the first eigenvector to obtain a first classification group, and matching in the basic user group matrix according to the second eigenvector to obtain a second classification group; acquiring corresponding recommended commodities according to the first classification group and the second classification group; and generating a background poster according to the recommended commodity.
According to the personalized background image generation method provided by the embodiment of the application, historical data of a plurality of users are collected; the historical data comprises access data and order data; establishing a first characteristic matrix according to the access data of each user, and establishing a second characteristic matrix according to the order data of each user; extracting a first feature vector of a target user from the first feature matrix, and extracting a second feature vector of the target user from the second feature matrix; matching in the basic user group matrix according to the first eigenvector to obtain a first classification group, and matching in the basic user group matrix according to the second eigenvector to obtain a second classification group; acquiring corresponding recommended commodities according to the first classification group and the second classification group; and generating a background poster according to the recommended commodity. According to the personalized background map generation method provided by the embodiment of the application, the first characteristic matrix used for representing the access habit of the user and the second characteristic matrix used for representing the shopping habit of the user are respectively established, the corresponding recommended commodities are obtained after the two characteristic matrices are combined, the accuracy of predicting favorite commodities of the target user is guaranteed, the advertisement promotion efficiency is improved, in addition, the obtained recommended commodities are displayed in a background poster generation mode, the acceptance of the recommended commodities by the user can be improved, and the advertisement promotion effect is further improved.
Drawings
Fig. 1 is an implementation environment diagram of a personalized background map generation method provided in an embodiment of the present application;
fig. 2 is a flowchart of a personalized background map generation method according to an embodiment of the present application;
fig. 3 is a flowchart of another personalized background map generation method provided in the embodiment of the present application;
fig. 4 is a flowchart of another personalized background map generation method provided in the embodiment of the present application;
fig. 5 is a flowchart of another personalized background map generation method provided in the embodiment of the present application;
fig. 6 is a flowchart of another personalized background map generation method provided in the embodiment of the present application;
fig. 7 is a flowchart of another personalized background map generation method provided in the embodiment of the present application;
fig. 8 is a flowchart of another personalized background map generation method provided in the embodiment of the present application;
fig. 9 is a flowchart of another personalized background map generation method provided in the embodiment of the present application;
fig. 10 is a block diagram of a personalized background map generation apparatus according to an embodiment of the present application;
fig. 11 is a block diagram of a computer device according to an embodiment of the present application.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.
The personalized background map generation method provided by the application can be applied to the application environment shown in fig. 1. Wherein the terminal device 102 communicates with the server 104 via a network. The terminal device 102 may be, but not limited to, various personal computers, notebook computers, smart phones, tablet computers, and portable wearable devices, and the server 104 may be implemented by an independent server or a server cluster formed by a plurality of servers.
Please refer to fig. 2, which shows a personalized background map generation method provided in this embodiment, and the method is applied to the server in fig. 1 as an example for description, and includes the following steps:
In an embodiment of the present application, the server 104 may be communicatively connected to a plurality of web servers, and specifically, each web server collects history data of a plurality of users on the terminal device, and then collects and caches the user history data collected by each web server through the kafka cluster, and finally stores the user history data in the server 104. The historical data of each user may include access data and order data of the user.
In another embodiment of the present application, the user history data may be collected and cached through any one of a message queue or a publish-subscribe message system in the Storm cluster, the Spark Streaming cluster and the scyihdb cluster.
And step 204, establishing a first characteristic matrix according to the access data of each user, and establishing a second characteristic matrix according to the order data of each user.
In one embodiment of the application, a plurality of collected users are used as one dimension, access data corresponding to each user are used as another dimension, a corresponding two-dimensional matrix is established, and the established two-dimensional matrix is used as a first feature matrix. The first feature matrix is used for representing the access habits of all users.
In an embodiment of the application, a plurality of collected users are used as one dimension, order data corresponding to each user is used as another dimension, a corresponding two-dimensional matrix is established, and the established two-dimensional matrix is used as a second feature matrix. The second feature matrix is used for representing the purchasing habits of all users.
And step 206, extracting a first feature vector of the target user from the first feature matrix, and extracting a second feature vector of the target user from the second feature matrix.
In an embodiment of the present application, an update time period is further set in the server, and the update time period may be set to 1 day, 1 week, or 1 month by default, or may be set to other times by a system administrator. When the updating time point in the updating time period is reached, the server sequentially extracts the users in the first characteristic matrix as target users and extracts corresponding first characteristic vectors; the server may also extract the users in the second feature matrix as target users in sequence, and extract corresponding second feature vectors.
In an embodiment of the application, the target user is a user currently accessing shopping software or a shopping webpage, when the target user accesses the shopping software or the shopping webpage through a terminal device, a server receives a login request carrying an identification code of the target user and sent by the terminal device, and the user corresponding to the identification code of the target user is taken as the target user.
Specifically, the access data corresponding to the target user identification code is extracted from the established first feature matrix to be used as a first feature vector, and the order data corresponding to the target user identification code is extracted from the established second feature matrix to be used as a second feature vector.
And 208, matching in the basic user group matrix according to the first eigenvector to obtain a first classification group, and matching in the basic user group matrix according to the second eigenvector to obtain a second classification group.
In an embodiment of the application, the basic user group matrix includes a plurality of standard classification groups and standard vectors corresponding to the standard classification groups, and the corresponding first classification group can be obtained by matching the first feature vector with each standard vector. Similarly, the second feature vector is matched with each standard vector to obtain a corresponding second classification group.
And step 210, acquiring corresponding recommended commodities according to the first classification group and the second classification group.
In an embodiment of the application, recommended commodities corresponding to each classification group are preset in the server, and the recommended commodities corresponding to the first classification group and the second classification group can be obtained through a mapping relation between each classification group and the recommended commodities.
In an embodiment of the application, the recommended commodities corresponding to each classification group may be a plurality of commodities, and after the commodities corresponding to the first classification group and the commodities corresponding to the second classification group are obtained, a plurality of commodities may be randomly selected from the commodities to serve as the recommended commodities corresponding to the first classification group and the second classification group; and according to the attribute of each commodity, screening a plurality of commodities to be used as recommended commodities corresponding to the first classification group and the second classification group.
And step 212, generating a background poster according to the recommended commodity.
In one embodiment of the present application, the generated background poster is used to display a recommended item corresponding to the target user.
In the method for generating the personalized background image, historical data of a plurality of users are collected; the historical data comprises access data and order data; establishing a first characteristic matrix according to the access data of each user, and establishing a second characteristic matrix according to the order data of each user; extracting a first feature vector of a target user from the first feature matrix, and extracting a second feature vector of the target user from the second feature matrix; matching in the basic user group matrix according to the first eigenvector to obtain a first classification group, and matching in the basic user group matrix according to the second eigenvector to obtain a second classification group; acquiring corresponding recommended commodities according to the first classification group and the second classification group; and generating a background poster according to the recommended commodity. According to the personalized background map generation method provided by the embodiment of the application, the first characteristic matrix used for representing the access habit of the user and the second characteristic matrix used for representing the shopping habit of the user are respectively established, the corresponding recommended commodities are obtained after the two characteristic matrices are combined, the accuracy of predicting favorite commodities of the target user is guaranteed, the advertisement promotion efficiency is improved, in addition, the obtained recommended commodities are displayed in a background poster generation mode, the acceptance of the recommended commodities by the user can be improved, and the advertisement promotion effect is further improved.
Please refer to fig. 3, which shows a flowchart of another personalized background map generation method provided in this embodiment, and the personalized background map generation method can be applied to the server 104 in the above implementation environment. On the basis of the embodiment shown in fig. 2, the step 204 may specifically include the following steps:
step 302, analyzing the access data of each user, wherein the access data comprises at least one access record carrying the information of the access commodity.
Specifically, the access data received by the server includes at least one access record, each access record carries a unique identification code of a user, and the access record corresponding to each user can be obtained by identifying the unique identification code. Each user corresponds to at least one access record, wherein each access record carries corresponding access commodity information.
In an embodiment of the application, the access record may be a record of the user entering the commodity detail interface, a record of the user clicking a commodity picture, or a record of the user staying in the commodity detail interface for a time longer than a preset time.
In one embodiment of the present application, the access merchandise information may be specific merchandise names, such as a specific model of mobile phone, a specific specification of food, and apparel.
And step 304, classifying at least one access record according to the preset mapping relation between the access commodity information and the commodity category.
Specifically, the commodity category may be clothing, books, food, home appliances, and the like, the mapping relationship between the access commodity information and the commodity category is a mapping relationship between a specific commodity and a category to which the commodity belongs, for example, the commodity category corresponding to down jackets, short sleeves, jeans, and the like is clothing, and the commodity category corresponding to televisions, refrigerators, air conditioners, and the like is home appliances.
In an embodiment of the application, the server extracts the access commodity information carried in the access record, searches for a corresponding commodity category according to the access commodity information, and classifies the access record according to the obtained commodity category.
And step 306, counting the number of access records of each user in each commodity category, and establishing a first feature matrix.
Specifically, one dimension of the first feature matrix is a plurality of users collected by the server, and the other dimension is the number of access records of each user in each commodity category.
In the method for generating the personalized background image, access data of each user are analyzed, and the access data comprise at least one access record carrying access commodity information; classifying at least one access record according to a preset mapping relation between the access commodity information and the commodity category; and counting the number of access records of each user in each commodity category, and establishing a first feature matrix. According to the personalized background map generation method provided by the embodiment of the application, the access records in the access data of each user are collected and analyzed, and each access record is classified to establish the first feature matrix capable of representing the interest value of the user under each commodity category, so that the access habits of the user on each commodity category can be accurately represented, and the accuracy of favorite commodity recommendation is improved.
Referring to fig. 4, a flowchart of another personalized background map generation method provided in this embodiment is shown, where the personalized background map generation method can be applied to the server 104 in the above implementation environment. On the basis of the embodiment shown in fig. 2, the step 204 may further include the following steps:
Specifically, the order data received by the server includes at least one shopping record, each shopping record carries a unique identification code of a user, and the shopping records corresponding to the users can be obtained by identifying the unique identification code, and each shopping record carries corresponding order commodity information.
In one embodiment of the present application, the shopping record may be a record of orders established for the user, or may be a record of orders that were successfully purchased.
And step 404, classifying at least one order record according to a preset mapping relation between order commodity information and commodity categories.
In an embodiment of the application, the server extracts the access commodity information carried in the shopping record, searches for a corresponding commodity category according to the access commodity information, and classifies the shopping record according to the obtained commodity category.
And 406, counting the order record quantity of each user in each commodity category, and establishing a second feature matrix.
Specifically, one dimension of the second feature matrix is a plurality of users collected by the server, and the other dimension is the number of order placing records of each user in each commodity category.
In the method for generating the personalized background image, order data of each user are analyzed, wherein the order data comprise at least one order record carrying order commodity information; classifying at least one order record according to a preset mapping relation between order commodity information and commodity categories; and counting the order record quantity of each user in each commodity category, and establishing a second feature matrix. According to the personalized background map generation method provided by the embodiment of the application, shopping records in order data of each user are collected and analyzed, and each shopping record is classified to establish the second feature matrix capable of representing interest values of the user under each commodity category, so that the purchasing habits of the user on each commodity category can be accurately represented, and the recommendation accuracy of favorite commodities is improved.
Please refer to fig. 5, which shows a flowchart of another personalized background map generation method provided in this embodiment, and the personalized background map generation method can be applied to the server 104 in the above implementation environment. On the basis of the embodiment shown in fig. 3, the step 306 may further include the following steps:
In one embodiment of the present application, one dimension of the first initial matrix is a plurality of users collected by the server, and the other dimension is the number of access records of each user under each commodity category.
And 504, performing noise reduction processing on the first initial matrix to remove the number of access records of which the access time is outside a preset calculation period and the access proportion is lower than an access threshold value, so as to obtain a first noise reduction matrix.
In an embodiment of the application, the server analyzes the access time of at least one access record corresponding to each user to obtain the number of access records of which the access time is outside the preset calculation period, and removes the number of access records of which the access time is outside the preset calculation period from the first initial matrix.
For example, the preset calculation period is 7 days, in the first initial matrix, the number of access records of the first user under the condition that the commodity category is food is 21, the server extracts the access time of each of the 21 access records of the first user under the condition that the category is food, counts the number of access records of which the access time is before 7 days, and if the number of access records is 6, the number of access records of the first user under the condition that the commodity category is food is 15 in the first noise reduction matrix after the noise reduction processing.
In one embodiment of the present application, if the number of access records of a user in a commodity category is lower than the access threshold, the number of access records of the user in the commodity category is zeroed.
For example, the access threshold is preset to 5, in the first initial matrix, the number of access records of the second user under the condition that the commodity category is food is 15, the server counts the number of access records of which the access time is 7 days ago, if the number is 11, the number of effective access records is 4, and if the number is lower than the access threshold, in the first denoising matrix after denoising processing, the number of access records of the second user under the condition that the commodity category is food is 0.
Specifically, a first noise reduction matrix is obtained by performing noise reduction processing on the first initial matrix, and the first noise reduction matrix is used as a first feature matrix.
In the method for generating the personalized background image, a first initial matrix is established according to the number of access records of each user in each commodity category; denoising the initial matrix to remove the number of access records of which the access time is outside a preset calculation period and the access proportion is lower than an access threshold value, and obtain a first denoising matrix; and generating a first feature matrix according to the first noise reduction matrix. According to the personalized background map generation method provided by the embodiment of the application, the first initial matrix is subjected to noise reduction processing, the number of access records with access time outside the preset calculation period and access proportion lower than the access threshold is removed, so that the first characteristic matrix is obtained, the access habits of users on various commodity categories can be represented more accurately, and the accuracy of favorite commodity recommendation is improved.
Please refer to fig. 6, which shows a flowchart of another personalized background map generation method provided in this embodiment, and the personalized background map generation method can be applied to the server 104 in the above implementation environment. On the basis of the embodiment shown in fig. 4, the step 406 may specifically include the following steps:
In one embodiment of the present application, one dimension of the first initial matrix is a plurality of users collected by the server, and the other dimension is the number of shopping records of each user in each commodity category.
And step 604, performing noise reduction processing on the initial matrix to remove the order record quantity with the purchase time outside the preset calculation period and the purchase ratio lower than the purchase threshold value, and obtaining a second noise reduction matrix.
In an embodiment of the application, the server analyzes the purchase time of at least one order record corresponding to each user to obtain the order record number of the purchase time outside the preset calculation period, and removes the order record number of the purchase time outside the preset calculation period from the second initial matrix.
For example, the preset calculation period is 7 days, in the second initial matrix, the number of the order records of the third user placed in the food in the commodity category is 21, the server extracts the order time of each of the 21 order records of the third user placed in the food in the category, and counts the number of the order records of which the order time is before 7 days, and if the number is 6, the number of the order records of the third user placed in the food in the commodity category is 15 in the second denoising matrix after denoising processing.
In one embodiment of the present application, if the number of order records placed by a user in a commodity category is lower than a purchase threshold, the number of order records placed by the user in the commodity category is zeroed.
For example, the purchase threshold is preset to be 3, in the second initial matrix, the order record number of the fourth user under the condition that the commodity category is food is 15, the server counts the order record number of the fourth user under the condition that the order time is 7 days ago, if the order record number is 14, the valid order record number is 1, and if the order record number is lower than the purchase threshold, in the second noise reduction matrix after the noise reduction processing, the order record number of the fourth user under the condition that the commodity category is food is 0.
And 606, generating a second feature matrix according to the second denoising matrix.
Specifically, a second noise reduction matrix is obtained by performing noise reduction processing on the second initial matrix, and the second noise reduction matrix is used as a second feature matrix.
In the method for generating the personalized background image, a second initial matrix is established according to the order record quantity of each user in each commodity category; denoising the initial matrix to remove the order record quantity of which the purchase time is out of the preset calculation period and the purchase ratio is lower than the purchase threshold value, and obtain a second denoising matrix; and generating a second feature matrix according to the second noise reduction matrix. According to the personalized background map generation method provided by the embodiment of the application, the second initial matrix is subjected to noise reduction processing, the order record quantity with the purchase time outside the preset calculation period and the access proportion lower than the purchase threshold is removed, so that the second characteristic matrix is obtained, the purchase habits of users for various commodity categories can be represented more accurately, and the accuracy of favorite commodity recommendation is improved.
The embodiment also provides another personalized background map generation method, which can be applied to the server 104 in the implementation environment described above. On the basis of the embodiment shown in fig. 5, the step 506 may further include the following steps:
and normalizing the number of the access records of each user in each commodity category in the first noise reduction matrix to obtain a first feature matrix.
In an embodiment of the application, the number of access records of each user in each commodity category is obtained, a maximum value and a non-zero minimum value are obtained, a value difference between the maximum value and the minimum value is further obtained, the number of access records of the user in each commodity category is divided by the obtained value difference, so that a numerical value of the user in each commodity category after normalization processing can be obtained, the numerical value represents a browsing interest value of the user in the corresponding commodity category, and a first feature matrix can be further obtained. For example, the number of access records of the fifth user in each commodity category is [ M ]1，M2，M3，…，Mn]Wherein M representsThe number of access records of the fifth user in a commodity category is recorded, and n represents the number of the commodity categories. The statistic in which the number of access technologies is the largest is denoted as MminThe minimum and non-zero number of access techniques is denoted as MmaxThe normalized value of the fifth user in each commodity category can be expressed as
According to the personalized background map generation method provided by the embodiment of the application, the influence caused by large access quantity difference among users can be avoided by performing normalization processing on the first noise reduction matrix, the access habits of the users on various commodity categories can be more accurately reflected, and the accuracy of favorite commodity recommendation is further improved.
The embodiment also provides another personalized background map generation method, which can be applied to the server 104 in the implementation environment described above. On the basis of the embodiment shown in fig. 6, the step 606 may further include the following steps:
and normalizing the order record quantity of each user in each commodity type in the second denoising matrix to obtain a second feature matrix.
In an embodiment of the application, the order record number of each user in each commodity category is obtained, a maximum value and a non-zero minimum value are obtained, a value difference between the maximum value and the minimum value is further obtained, the order record number of the user in each commodity category is divided by the obtained value difference, so that a numerical value of the user in each commodity category after normalization processing can be obtained, the numerical value represents a purchase interest value of the user in the corresponding commodity category, and a second feature matrix can be further obtained. For example, the sixth user records the number of orders in each commodity category as [ B ]1，B2，B3，…，Bn]Wherein, B represents the order record number of the sixth user in a certain commodity category, and n represents the number of the commodity category. The statistics in which the technical number of orders is the most is represented as BminThe technical quantity of the order is the bestA few and non-zero is denoted as BmaxThe normalized value of the sixth user in each commodity category may be represented as
According to the personalized background map generation method provided by the embodiment of the application, the second denoising matrix is subjected to normalization processing, so that the influence caused by large order quantity difference among users can be avoided, the purchasing habits of the users on various commodity categories can be more accurately reflected, and the accuracy of favorite commodity recommendation is further improved.
Referring to fig. 7, a flowchart of another personalized background map generation method provided in this embodiment is shown, where the personalized background map generation method can be applied to the server 104 in the above implementation environment. On the basis of the embodiment shown in fig. 2, the step 208 may specifically include the following steps:
In one embodiment of the present application, the basic user group matrix includes a plurality of standard classification groups and standard vectors corresponding to the standard classification groups, the similarity between the first feature vector and each standard vector is calculated, and the standard vector with the highest similarity is used as the first standard vector matched with the first feature vector. And taking the standard classification group corresponding to the first standard vector as a first classification group corresponding to the target user, wherein the first classification group is used for representing the standard classification group to which the target user should belong in the view of access habits.
And step 708, taking the standard classification group corresponding to the second standard vector as a second classification group.
In one embodiment of the present application, the similarity between the second feature vector and each standard vector is calculated, and the standard vector with the highest similarity is used as the second standard vector matched with the second feature vector. And taking the standard classification group corresponding to the second standard vector as a second classification group corresponding to the target user, wherein the second classification group is used for representing the standard classification group to which the target user belongs in the aspect of shopping habits.
Specifically, the similarity between the feature vector and the standard vector can be calculated in the following manner if there is a feature vector of [0.1,0.6,0.4,0.8 ]]The standard vector of the A-standard classification population is [0.2,0.3,0.8,0.1 ]]The standard vector of the B-standard classification population is [0.2,0.8,0.1, 0.3 ]]The standard vector of the C standard classification population is [0.8,0.2,0.6,0.4 ]]. Wherein the similarity of the feature vector and the A-standard classification population can be determined by
Calculating to obtain that the similarity between the feature vector and the A-standard classification group is 0.57, and by using the same calculation method, the similarity between the feature vector and the B-standard classification group is 0.69, the similarity between the feature vector and the C-standard classification group is 0.52, and the higher the similarity is, the closer the corresponding standard vector and the feature vector are, so that the standard vector of the B-standard classification group is the standard vector with the highest matching degree, and the standard classification group corresponding to the feature vector is the B-standard classification group.
According to the personalized background image generation method provided by the embodiment of the application, the standard classification group corresponding to the feature vector is obtained by matching the feature vector with each standard vector and calculating the similarity, so that the accuracy of the matching process is improved, the accuracy of predicting the favorite commodities of the target user is further ensured, and the advertisement promotion efficiency is improved.
Please refer to fig. 8, which shows a flowchart of another personalized background map generation method provided in this embodiment, and the personalized background map generation method can be applied to the server 104 in the above implementation environment. On the basis of the embodiment shown in fig. 2, the step 210 may further include the following steps:
In an embodiment of the present application, each of the plurality of standard classification groups included in the basic user group matrix includes a corresponding prediction result, that is, each of the standard classification groups is provided with at least one favorite product of the group and an interest value corresponding to each favorite product. By searching the first classification group and the second classification matrix in the basic user group matrix, at least one first classification commodity corresponding to the first classification group and an interest value corresponding to each first classification commodity, and at least one second classification commodity corresponding to the second classification group and an interest value corresponding to each second classification commodity can be obtained.
In an embodiment of the application, the obtained at least one first classified commodity and at least one second classified commodity are ranked according to the interest value, and one or more commodities with the highest interest value are used as the recommended commodities.
In an embodiment of the present application, a numerical value of a category of the favorite product in the corresponding standard vector may be used as the interest value of the favorite product.
According to the personalized background map generation method provided by the embodiment of the application, the commodity recommendation results of the target user in the access angle and the purchase angle are fused, so that a more accurate commodity recommendation result can be obtained, the prediction accuracy of the favorite commodity of the target user is ensured, and the advertisement promotion efficiency is improved.
Please refer to fig. 9, which shows a flowchart of another personalized background map generation method provided in this embodiment, and the personalized background map generation method can be applied to the server 104 in the above implementation environment. On the basis of the embodiment shown in fig. 2, the step 212 may further include the following steps:
step 902, a merchandise pattern and a coupon pattern corresponding to the recommended merchandise are obtained.
And 904, generating a background poster according to the commodity pattern and the coupon pattern.
In an embodiment of the application, the server searches a product image corresponding to the recommended product in a database, and the product pattern may be a rendering image of the recommended product or a real shooting image of the recommended product.
In an embodiment of the present application, the coupon pattern may be a coupon pattern corresponding to the recommended product, may also be a coupon pattern corresponding to a product category corresponding to the recommended product, and may also be a coupon pattern corresponding to a manufacturer to which the recommended product belongs. In particular, coupon patterns include, but are not limited to, discount coupons, and gift coupons.
In an embodiment of the application, after the server obtains the commodity patterns and the coupon patterns, the corresponding commodity links and coupon getting links are attached to the corresponding patterns, and the picture renderer renders the links corresponding to the patterns and the links corresponding to the patterns onto the same background picture to obtain the background poster. The background poster is stored in a database and establishes a mapping relationship with the target user.
According to the personalized background image generation method provided by the embodiment of the application, the obtained recommended commodities are displayed in a background poster generation mode, the acceptance of users to the recommended commodities can be improved, and the advertisement promotion effect is further improved. And moreover, the convenience of purchasing commodities by the user is improved by adding the corresponding jump links at the corresponding commodity patterns and the corresponding coupon patterns.
The embodiment also provides another personalized background map generation method, which can be applied to the server 104 in the implementation environment described above. On the basis of the embodiment shown in fig. 2, the method may further include the following steps:
and receiving a login request of a target user on the terminal equipment. And pushing the background poster to the terminal equipment.
In one embodiment of the application, each time a user accesses shopping software or a shopping webpage in a terminal device, a server receives a login request which is sent by the terminal device and carries a target user identification code, and searches a corresponding background poster in a database according to the target user identification code. And if the corresponding background poster is found, pushing the background poster to terminal equipment logged by the user, and displaying the background poster on a webpage or a software specific interface in a background picture mode. And if the corresponding background poster is not found, pushing the default background poster to terminal equipment logged by the user, and displaying the background poster on a webpage or a software specific interface in a background picture mode.
According to the personalized background picture generation method provided by the embodiment of the application, the corresponding background poster is immediately searched and sent to the terminal equipment logged by the user by receiving the login request of the user, and the background poster is displayed on a webpage or a software specific interface in a background picture mode, so that the instantaneity of advertisement promotion can be ensured, and the advertisement promotion efficiency is further improved.
It should be understood that, although the steps in the above-described flowcharts are shown in order as indicated by the arrows, the steps are not necessarily performed in order as indicated by the arrows. The steps are not performed in the exact order shown and described, and may be performed in other orders, unless explicitly stated otherwise. Moreover, at least a portion of the steps in the above-described flowcharts may include multiple sub-steps or multiple stages, which are not necessarily performed at the same time, but may be performed at different times, and the order of performing the sub-steps or the stages is not necessarily sequential, but may be performed alternately or alternatingly with other steps or at least a portion of the sub-steps or stages of other steps.
Referring to fig. 10, a block diagram of a personalized background image generation apparatus 1000 according to an embodiment of the present application is shown. As shown in fig. 10, the personalized background map generation system 1000 may include: a data collection module 1001, a feature matrix establishing module 1002, a feature vector extraction module 1003, a matching module 1004, a recommendation module 1005 and a poster generation module 1006, wherein:
a data collection module 1001 for collecting history data of a plurality of users; the historical data includes access data and order data.
The feature matrix creating module 1002 is configured to create a first feature matrix according to the access data of each user, and create a second feature matrix according to the order data of each user.
A feature vector extraction module 1003, configured to extract a first feature vector of a target user in the first feature matrix, and extract a second feature vector of the target user in the second feature matrix.
The matching module 1004 is configured to perform matching in the basic user group matrix according to the first feature vector to obtain a first classification group, and perform matching in the basic user group matrix according to the second feature vector to obtain a second classification group.
The recommending module 1005 is configured to obtain the corresponding recommended commodity according to the first classification group and the second classification group.
And a poster generation module 1006, configured to generate a background poster according to the recommended product.
In an embodiment of the present application, the feature matrix establishing module 1002 is specifically configured to: analyzing access data of each user, wherein the access data comprises at least one access record carrying access commodity information; classifying at least one access record according to a preset mapping relation between the access commodity information and the commodity category; and counting the number of access records of each user in each commodity category, and establishing a first feature matrix.
In an embodiment of the present application, the feature matrix establishing module 1002 is further specifically configured to: analyzing order data of each user, wherein the order data comprises at least one order record carrying order commodity information; classifying at least one order record according to a preset mapping relation between order commodity information and commodity categories; and counting the order record quantity of each user in each commodity category, and establishing a second feature matrix.
In an embodiment of the present application, the feature matrix establishing module 1002 is further specifically configured to: establishing a first initial matrix according to the number of access records of each user in each commodity category; denoising the initial matrix to remove the number of access records of which the access time is outside a preset calculation period and the access proportion is lower than an access threshold value, and obtain a first denoising matrix; and generating a first feature matrix according to the first noise reduction matrix.
In an embodiment of the present application, the feature matrix establishing module 1002 is further specifically configured to: establishing a second initial matrix according to the order record quantity of each user in each commodity category; denoising the initial matrix to remove the order record quantity of which the purchase time is out of the preset calculation period and the purchase ratio is lower than the purchase threshold value, and obtain a second denoising matrix; and generating a second feature matrix according to the second noise reduction matrix.
In an embodiment of the present application, the feature matrix establishing module 1002 is further specifically configured to: and normalizing the number of the access records of each user in each commodity category in the first noise reduction matrix to obtain a first feature matrix.
In an embodiment of the present application, the feature matrix establishing module 1002 is further specifically configured to: and normalizing the order record quantity of each user in each commodity type in the second denoising matrix to obtain a second feature matrix.
In an embodiment of the application, the basic user group matrix includes a plurality of standard classification groups and standard vectors corresponding to the standard classification groups, and the matching module 1004 is specifically configured to: calculating the similarity between the first feature vector and each standard vector, and taking the standard vector with the highest similarity as the first standard vector matched with the first feature vector; taking a standard classification group corresponding to the first standard vector as a first classification group; calculating the similarity between the second feature vector and each standard vector, and taking the standard vector with the highest similarity as the second standard vector matched with the second feature vector; and taking the standard classification group corresponding to the second standard vector as a second classification group.
In an embodiment of the present application, the recommending module 1005 is specifically configured to: obtaining a first prediction result corresponding to the first classified group, wherein the first prediction result comprises at least one first classified commodity and an interest value corresponding to each first classified commodity; obtaining a second prediction result corresponding to the second classified group, wherein the second prediction result comprises at least one second classified commodity and an interest value corresponding to each second classified commodity; and acquiring the recommended commodity according to the interest value corresponding to each first classified commodity and the interest value corresponding to each second classified commodity.
In an embodiment of the present application, the poster generation module 1006 is specifically configured to: acquiring a commodity pattern and a coupon pattern corresponding to the recommended commodity; and generating a background poster according to the commodity pattern and the coupon pattern.
In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as shown in fig. 11. The computer device includes a processor, a memory, a network interface, and a database connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a personalized background map generation method.
Those skilled in the art will appreciate that the architecture shown in fig. 11 is merely a block diagram of some of the structures associated with the disclosed aspects and is not intended to limit the computing devices to which the disclosed aspects apply, as particular computing devices may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.
In one embodiment, a computer device is provided, comprising a memory and a processor, the memory having a computer program stored therein, the processor implementing the following steps when executing the computer program:
collecting historical data of a plurality of users; the historical data comprises access data and order data; establishing a first characteristic matrix according to the access data of each user, and establishing a second characteristic matrix according to the order data of each user; extracting a first feature vector of a target user from the first feature matrix, and extracting a second feature vector of the target user from the second feature matrix; matching in the basic user group matrix according to the first eigenvector to obtain a first classification group, and matching in the basic user group matrix according to the second eigenvector to obtain a second classification group; acquiring corresponding recommended commodities according to the first classification group and the second classification group; and generating a background poster according to the recommended commodity.
In one embodiment, a computer-readable storage medium is provided, having a computer program stored thereon, which when executed by a processor, performs the steps of:
collecting historical data of a plurality of users; the historical data comprises access data and order data; establishing a first characteristic matrix according to the access data of each user, and establishing a second characteristic matrix according to the order data of each user; extracting a first feature vector of a target user from the first feature matrix, and extracting a second feature vector of the target user from the second feature matrix; matching in the basic user group matrix according to the first eigenvector to obtain a first classification group, and matching in the basic user group matrix according to the second eigenvector to obtain a second classification group; acquiring corresponding recommended commodities according to the first classification group and the second classification group; and generating a background poster according to the recommended commodity.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in the embodiments provided herein may include non-volatile and/or volatile memory, among others. Non-volatile memory can include read-only memory (ROM), Programmable ROM (PROM), Electrically Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDRSDRAM), Enhanced SDRAM (ESDRAM), Synchronous Link DRAM (SLDRAM), Rambus Direct RAM (RDRAM), direct bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).
The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.
The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present patent shall be subject to the appended claims.
Claims (11)
1. A background map generation method, characterized in that the method comprises:
collecting historical data of a plurality of users; the historical data comprises access data and order data;
establishing a first characteristic matrix according to the access data of each user, and establishing a second characteristic matrix according to the order data of each user;
extracting a first feature vector of a target user from the first feature matrix, and extracting a second feature vector of the target user from the second feature matrix;
matching in a basic user group matrix according to the first eigenvector to obtain a first classification group, and matching in the basic user group matrix according to the second eigenvector to obtain a second classification group;
acquiring corresponding recommended commodities according to the first classification group and the second classification group;
and generating a background poster according to the recommended commodity.
2. The method of claim 1, wherein the establishing a first feature matrix based on the access data of each of the users comprises:
analyzing access data of each user, wherein the access data comprises at least one access record carrying access commodity information;
classifying the at least one access record according to a preset mapping relation between the access commodity information and the commodity category;
and counting the number of access records of each user in each commodity category, and establishing the first feature matrix.
3. The method of claim 1, wherein said building a second feature matrix from order data of each of said users comprises: analyzing order data of each user, wherein the order data comprises at least one order record carrying order commodity information;
classifying the at least one order record according to a preset mapping relation between order commodity information and commodity categories;
and counting the order record quantity of each user in each commodity category, and establishing the second feature matrix.
4. The method of claim 2, wherein said counting the number of access records of each of said users in each of said categories of merchandise, and establishing said first feature matrix comprises:
establishing a first initial matrix according to the number of access records of each user in each commodity category;
denoising the initial matrix to remove the number of access records of which the access time is outside a preset calculation period and the access proportion is lower than an access threshold value, and obtain a first denoising matrix;
and generating the first feature matrix according to the first noise reduction matrix.
5. The method of claim 3, wherein said counting the number of order records of each of said users in each of said categories of goods, and establishing said second feature matrix comprises:
establishing a second initial matrix according to the order record quantity of each user in each commodity category;
denoising the initial matrix to remove the order record quantity of which the purchase time is out of the preset calculation period and the purchase ratio is lower than the purchase threshold value, and obtain a second denoising matrix;
and generating the second feature matrix according to the second noise reduction matrix.
6. The method of claim 4, wherein the generating the first feature matrix from the first noise reduction matrix comprises:
and normalizing the number of the access records of each user in each commodity category in the first noise reduction matrix to obtain the first feature matrix.
7. The method of claim 5, wherein the generating the second feature matrix from the second noise reduction matrix comprises:
and normalizing the order record quantity of each user in each commodity category in the second denoising matrix to obtain the second feature matrix.
8. The method of claim 1, wherein the basic user population matrix comprises a plurality of standard classification populations and standard vectors corresponding to the standard classification populations, and wherein the matching in the basic user population matrix according to the first eigenvector to obtain a first classification population and the matching in the basic user population matrix according to the second eigenvector to obtain a second classification population comprises:
calculating the similarity between the first feature vector and each standard vector, and taking the standard vector with the highest similarity as the first standard vector matched with the first feature vector;
taking a standard classification group corresponding to the first standard vector as the first classification group;
calculating the similarity between the second feature vector and each standard vector, and taking the standard vector with the highest similarity as the second standard vector matched with the second feature vector;
and taking the standard classification group corresponding to the second standard vector as the second classification group.
9. The method of claim 8, wherein the obtaining corresponding recommended goods according to the first and second taxonomic groups comprises:
obtaining a first prediction result corresponding to the first classified group, wherein the first prediction result comprises at least one first classified commodity and an interest value corresponding to each first classified commodity;
obtaining a second prediction result corresponding to the second classified group, wherein the second prediction result comprises at least one second classified commodity and an interest value corresponding to each second classified commodity;
and acquiring the recommended commodity according to the interest value corresponding to each first classified commodity and the interest value corresponding to each second classified commodity.
10. The method of claim 1, wherein generating the background poster from the recommended article comprises:
acquiring a commodity pattern and a coupon pattern corresponding to the recommended commodity;
and generating a background poster according to the commodity pattern and the coupon pattern.
11. A background map generation apparatus, characterized in that the apparatus comprises:
the data collection module is used for collecting historical data of a plurality of users; the historical data comprises access data and order data;
the characteristic matrix establishing module is used for establishing a first characteristic matrix according to the access data of each user and establishing a second characteristic matrix according to the order data of each user;
the feature vector extraction module is used for extracting a first feature vector of a target user from the first feature matrix and extracting a second feature vector of the target user from the second feature matrix;
the matching module is used for matching in a basic user group matrix according to the first characteristic vector to obtain a first classification group and matching in the basic user group matrix according to the second characteristic vector to obtain a second classification group;
the recommending module is used for acquiring corresponding recommended commodities according to the first classification group and the second classification group;
and the poster generating module is used for generating a background poster according to the recommended commodity.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910836136.9A CN110675217A (en) | 2019-09-05 | 2019-09-05 | Personalized background image generation method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910836136.9A CN110675217A (en) | 2019-09-05 | 2019-09-05 | Personalized background image generation method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110675217A true CN110675217A (en) | 2020-01-10 |
Family
ID=69076011
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910836136.9A Pending CN110675217A (en) | 2019-09-05 | 2019-09-05 | Personalized background image generation method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110675217A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113643075A (en) * | 2021-07-04 | 2021-11-12 | 杭州拼便宜网络科技有限公司 | Intelligent poster generation method and device based on machine learning and storage medium |
| CN114493739A (en) * | 2021-12-07 | 2022-05-13 | 珠海格力电器股份有限公司 | A weather information-based recommendation method, system, and electronic device |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101916279A (en) * | 2010-08-16 | 2010-12-15 | 刘文萍 | Plant species query system and method based on image analysis technology |
| CN102306182A (en) * | 2011-08-30 | 2012-01-04 | 西华大学 | Method for excavating user interest based on conceptual semantic background image |
| US20140279205A1 (en) * | 2013-03-14 | 2014-09-18 | United Parcel Service Of America, Inc. | Date reminder systems and related methods |
| CN105989004A (en) * | 2015-01-27 | 2016-10-05 | 阿里巴巴集团控股有限公司 | Information releasing pretreatment method and device |
| US20170091805A1 (en) * | 2014-06-16 | 2017-03-30 | Huawei Technologies Co., Ltd. | Advertisement Recommendation Method and Advertisement Recommendation Server |
| CN108805594A (en) * | 2017-04-27 | 2018-11-13 | 北京京东尚科信息技术有限公司 | Information-pushing method and device |
| CN109934721A (en) * | 2019-01-18 | 2019-06-25 | 深圳壹账通智能科技有限公司 | Finance product recommended method, device, equipment and storage medium |
-
2019
- 2019-09-05 CN CN201910836136.9A patent/CN110675217A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101916279A (en) * | 2010-08-16 | 2010-12-15 | 刘文萍 | Plant species query system and method based on image analysis technology |
| CN102306182A (en) * | 2011-08-30 | 2012-01-04 | 西华大学 | Method for excavating user interest based on conceptual semantic background image |
| US20140279205A1 (en) * | 2013-03-14 | 2014-09-18 | United Parcel Service Of America, Inc. | Date reminder systems and related methods |
| US20170091805A1 (en) * | 2014-06-16 | 2017-03-30 | Huawei Technologies Co., Ltd. | Advertisement Recommendation Method and Advertisement Recommendation Server |
| CN105989004A (en) * | 2015-01-27 | 2016-10-05 | 阿里巴巴集团控股有限公司 | Information releasing pretreatment method and device |
| CN108805594A (en) * | 2017-04-27 | 2018-11-13 | 北京京东尚科信息技术有限公司 | Information-pushing method and device |
| CN109934721A (en) * | 2019-01-18 | 2019-06-25 | 深圳壹账通智能科技有限公司 | Finance product recommended method, device, equipment and storage medium |
Non-Patent Citations (1)
| Title |
|---|
| 张海建: "分布式矩阵分解算法在推荐系统中的研究与应用", 《科技通报》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113643075A (en) * | 2021-07-04 | 2021-11-12 | 杭州拼便宜网络科技有限公司 | Intelligent poster generation method and device based on machine learning and storage medium |
| CN114493739A (en) * | 2021-12-07 | 2022-05-13 | 珠海格力电器股份有限公司 | A weather information-based recommendation method, system, and electronic device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107424043B (en) | Product recommendation method and device and electronic equipment | |
| CN108665329B (en) | Commodity recommendation method based on user browsing behavior | |
| CN113077317B (en) | Item recommendation method, device, equipment and storage medium based on user data | |
| CN111080398B (en) | Commodity recommendation method, commodity recommendation device, computer equipment and storage medium | |
| CN109783730A (en) | Products Show method, apparatus, computer equipment and storage medium | |
| CN109493199A (en) | Products Show method, apparatus, computer equipment and storage medium | |
| CN110008397B (en) | Recommendation model training method and device | |
| CN107247786A (en) | Method, device and server for determining similar users | |
| CN106557480B (en) | Method and device for realizing query rewriting | |
| CN112001754A (en) | User portrait generation method, device, equipment and computer readable medium | |
| CN109168047B (en) | Video recommendation method and device, server and storage medium | |
| CN118628214B (en) | Personalized clothing recommendation method and system for electronic commerce platform based on artificial intelligence | |
| CN110674391B (en) | Product data pushing method and system based on big data and computer equipment | |
| CN109409928A (en) | A kind of material recommended method, device, storage medium, terminal | |
| CN113032668A (en) | Product recommendation method, device and equipment based on user portrait and storage medium | |
| CN112149003B (en) | Commodity community recommendation method and device and computer equipment | |
| CN112241489A (en) | Information pushing method and device, readable storage medium and computer equipment | |
| CN109801101A (en) | Label determines method, apparatus, computer equipment and storage medium | |
| CN112837108A (en) | Information processing method, device and electronic device | |
| CN110675217A (en) | Personalized background image generation method and device | |
| CN115496523A (en) | Advertisement publishing method, computer readable storage medium and electronic device | |
| CN119311958A (en) | User analysis method, device, equipment and medium based on big model | |
| CN112036987B (en) | Method and device for determining recommended commodity | |
| CN114399352A (en) | Information recommendation method and device, electronic equipment and storage medium | |
| CN113271325B (en) | Information push method, device, electronic device and computer readable medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200110 |