CN111681086A - Commodity recommendation method and device, computer equipment and readable storage medium - Google Patents
Commodity recommendation method and device, computer equipment and readable storage medium Download PDFInfo
- Publication number
- CN111681086A CN111681086A CN202010550930.XA CN202010550930A CN111681086A CN 111681086 A CN111681086 A CN 111681086A CN 202010550930 A CN202010550930 A CN 202010550930A CN 111681086 A CN111681086 A CN 111681086A
- Authority
- CN
- China
- Prior art keywords
- commodity
- recommended
- target
- user
- operation data
- 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
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]
- G06Q30/0631—Item recommendations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
-
- 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
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
- G06Q30/0202—Market predictions or forecasting for commercial activities
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Development Economics (AREA)
- Theoretical Computer Science (AREA)
- Strategic Management (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Artificial Intelligence (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Health & Medical Sciences (AREA)
- Game Theory and Decision Science (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application provides a commodity recommendation method, a commodity recommendation device, computer equipment and a readable storage medium, relates to the technical field of electronic commerce, and is applied to the computer equipment, wherein the computer equipment stores operation data of a plurality of users on a target shopping platform, and the commodity recommendation method comprises the following steps: acquiring operation data of a target user on a target shopping platform; the method comprises the steps of obtaining operation data of a plurality of other users on a target shopping platform, wherein the other users are users except a target user in the plurality of users; obtaining interested commodities of the target user through collaborative filtering calculation according to the operation data of the target user on the target shopping platform and the operation data of each other user on the target shopping platform; acquiring a to-be-recommended file and extracting a to-be-recommended commodity from the to-be-recommended file; when the commodities to be recommended are the same as the interested commodities, the file to be recommended is recommended to the target user as the recommended file, and the commodities can be accurately recommended to the target user.
Description
Technical Field
The application relates to the technical field of electronic commerce, in particular to a commodity recommendation method, a commodity recommendation device, computer equipment and a readable storage medium.
Background
At present, operators of merchants can write paperwork to recommend commodities. However, a merchant usually has a large number of customers, if the documents written by operators are pushed to each customer, because the pertinence is not strong, the relevant documents of the commodities which the customers do not have interest in may be recommended, and in the past, the stickiness of the user to the merchant is reduced, even the trouble of the customers is increased, and finally the merchant is not concerned any more, the recommendation of the commodities cannot be accurately performed, and the benefit of the merchant is damaged.
In view of this, it is necessary for those skilled in the art to provide a more accurate commodity recommendation scheme.
Disclosure of Invention
The application provides a commodity recommendation method, a commodity recommendation device, computer equipment and a readable storage medium.
The embodiment of the application can be realized as follows:
in a first aspect, an embodiment provides a commodity recommendation method, which is applied to a computer device, where the computer device stores operation data of multiple users on a target shopping platform, and the method includes:
acquiring operation data of a target user on the target shopping platform;
acquiring operation data of a plurality of other users in the target shopping platform, wherein the other users are users except the target user in the plurality of users;
obtaining interested commodities of the target user through collaborative filtering calculation according to the operation data of the target user on the target shopping platform and the operation data of each other user on the target shopping platform;
acquiring a to-be-recommended file and extracting a to-be-recommended commodity from the to-be-recommended file;
and when the to-be-recommended commodity is the same as the interested commodity, recommending the to-be-recommended file to the target user as a recommended file.
In an optional embodiment, the document to be recommended includes at least one commodity title, the commodity title has a corresponding relationship with a commodity, the computer device stores a commodity title word segmentation word bank, and the commodity title word segmentation word bank includes a plurality of commodity titles;
the step of extracting the goods to be recommended from the file to be recommended comprises the following steps:
segmenting words of the to-be-recommended case through the commodity title segmentation word bank to obtain at least one commodity title of the to-be-recommended case;
and taking the commodity corresponding to the commodity title of which the occurrence frequency exceeds a preset threshold value in the at least one commodity title as the commodity to be recommended.
In an optional embodiment, the commodity name includes a commodity name and/or a commodity short name and/or a commodity nickname, and the commodity name participle thesaurus further includes a corresponding relationship among the commodity name, the commodity short name and the commodity nickname;
the step of performing word segmentation on the to-be-recommended case through the commodity title word segmentation word bank to obtain at least one commodity title of the to-be-recommended case comprises the following steps of:
performing word segmentation on the to-be-recommended case through the commodity title word segmentation word bank to obtain at least one commodity name and/or commodity short name and/or commodity nickname;
the step of taking the commodity corresponding to the commodity title of which the occurrence frequency exceeds a preset threshold value in the at least one commodity title as the commodity to be recommended comprises the following steps:
and according to the corresponding relation among the commodity names, the short commodity names and the nicknames of the commodities, the commodities corresponding to the commodity names and/or the short commodity names and/or the nicknames of the commodities, the occurrence times of which exceed a preset threshold value, are taken as the commodities to be recommended.
In an optional embodiment, the document to be recommended further includes a link to the goods to be recommended;
the step of extracting the goods to be recommended from the file to be recommended comprises the following steps:
extracting the commodity link to be recommended from the file to be recommended;
and analyzing the commodity link to be recommended to obtain the commodity to be recommended corresponding to the commodity link to be recommended.
In an optional embodiment, the step of obtaining the target user's interest goods through collaborative filtering calculation according to the operation data of the target user on the target shopping platform and the operation data of each of the other users on the target shopping platform includes:
calculating the user similarity of the target user and each other user according to the operation data of the target user on the target shopping platform and the operation data of each other user on the target shopping platform;
according to the user similarity between the target user and each of the other users, determining a reference user with high similarity to the target user from the other users;
and calculating to obtain the interested commodity of the target user according to the operation data of the reference user on the target shopping platform and the operation data of the target user on the target shopping platform.
In an alternative embodiment, the operational data of the reference user at the target shopping platform includes reference goods;
the step of calculating the interest goods of the target user according to the operation data of the reference user on the target shopping platform and the operation data of the target user on the target shopping platform comprises the following steps:
calculating the grade of the reference user on the reference commodity in the operation data of the target shopping platform through an entropy weight method according to the operation data of the reference user on the target shopping platform;
calculating to obtain an interest value of the target user to the reference commodity through a preset formula according to the score of the reference commodity and the user similarity of the reference user and the target user;
and when the interest value of the target user to the reference commodity exceeds a preset interest threshold value, taking the reference commodity as the interest commodity of the target user.
In an alternative embodiment, the score of the reference commodity is calculated by the following formula:
wherein s isiThe grade of the reference commodity i is determined, k is the operation type of the reference user on the reference commodity i on the target shopping platform, wkIs the weight of the operation type k, numkIs the number of executions of the operation type k.
In a second aspect, an embodiment provides an article recommendation apparatus applied to a computer device, where the computer device stores operation data of a plurality of users on a target shopping platform, and the apparatus includes:
the acquisition module is used for acquiring the operation data of the target user on the target shopping platform; acquiring operation data of a plurality of other users in the target shopping platform, wherein the other users are users except the target user in the plurality of users;
the calculation module is used for obtaining interested commodities of the target user through collaborative filtering calculation according to the operation data of the target user on the target shopping platform and the operation data of each other user on the target shopping platform;
the recommendation module is used for acquiring a to-be-recommended file and extracting a to-be-recommended commodity from the to-be-recommended file; and when the to-be-recommended commodity is the same as the interested commodity, recommending the to-be-recommended file to the target user as a recommended file.
In a third aspect, an embodiment provides a computer device, which includes a processor and a non-volatile memory storing computer instructions, and when the computer instructions are executed by the processor, the computer device executes the item recommendation method according to any one of the foregoing embodiments.
In a fourth aspect, an embodiment provides a readable storage medium, which includes a computer program, where the computer program is executed to control a computer device in the readable storage medium to execute the product recommendation method according to any one of the foregoing embodiments.
The beneficial effects of the embodiment of the application include, for example:
by adopting the commodity recommendation method, the commodity recommendation device, the computer equipment and the readable storage medium, the operation data of the target user on the target shopping platform is obtained, and the operation data of a plurality of other users on the target shopping platform is referred by the other users; then obtaining interested commodities of the target user through collaborative filtering calculation according to the operation data of the target user on the target shopping platform and the operation data of each other user on the target shopping platform; then obtaining a document to be recommended and extracting a commodity to be recommended from the document to be recommended; and when the commodity to be recommended is the same as the interested commodity, recommending the file to be recommended to the target user as a recommended file, and accurately recommending the commodity of the target user.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are required to be used in the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present application and therefore should not be considered as limiting the scope, and for those skilled in the art, other related drawings can be obtained from the drawings without inventive effort.
FIG. 1 is a schematic flow chart illustrating steps of a commodity recommendation method according to the present application;
FIG. 2 is a schematic flow chart illustrating steps of another commodity recommendation method provided in the present application;
FIG. 3 is a flowchart illustrating steps of another method for recommending merchandise according to the present application;
FIG. 4 is a flowchart illustrating steps of another merchandise recommendation method implemented and provided herein;
FIG. 5 is a flowchart illustrating steps of another merchandise recommendation method implemented and provided herein;
fig. 6 is a block diagram schematically illustrating a structure of a product recommendation device according to an embodiment of the present application;
FIG. 7 is a block diagram illustrating a computer device in accordance with an embodiment of the present invention.
Icon: 100-a computer device; 110-a commodity recommendation device; 1101-an acquisition module; 1102-a calculation module; 1103-recommendation module; 111-a memory; 112-a processor; 113-communication unit.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are some embodiments of the present application, but not all embodiments. The components of the embodiments of the present application, generally described and illustrated in the figures herein, can be arranged and designed in a wide variety of different configurations.
Thus, the following detailed description of the embodiments of the present application, presented in the accompanying drawings, is not intended to limit the scope of the claimed application, but is merely representative of selected embodiments of the application. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
It should be noted that the features of the embodiments of the present application may be combined with each other without conflict.
Currently, in order to increase the sales of goods, a merchant hires an operator to write a relevant case for recommending goods and pushes the case to a user of the merchant. The number of users of a merchant is generally huge (for example, tens of thousands of users), and when each document written by an operator is pushed to each user indiscriminately, the user receives the document which is not interested by the user, so that not only the sales amount of the merchant cannot be increased, but also the user is troubled, and the user may not pay attention to the merchant any more. In the prior art, commodity recommendation cannot be accurately performed on a single user, so that the user viscosity is reduced, and finally the user runs away. Based on this, an embodiment of the present application provides a product recommendation method, which is applied to a computer device, where the computer device stores operation data of a plurality of users on a target shopping platform, as shown in fig. 1, the embodiment of the present application includes steps 201 to 205.
In order to make targeted recommendation to the user, operation data of the target user on a target shopping platform may be obtained first, where the target platform may refer to a specific shopping APP (Application), or may refer to a shopping website, and is not limited herein. The acquired operation data may include the name of the product operated by the user, or may include what kind of operation (e.g., click, purchase, etc.) is specifically performed on the product.
The plurality of other users are users except the target user in the plurality of users.
In order to accurately recommend the target users, the operation data of other users on the same target shopping platform can be acquired, so that the accuracy of the subsequent reference for determining the interest commodities of the target users is ensured.
And step 203, obtaining the interested commodities of the target user through collaborative filtering calculation according to the operation data of the target user on the target shopping platform and the operation data of each other user on the target shopping platform.
In the embodiment of the application, the interested commodities of the target user are determined in a collaborative filtering manner, and compared with the traditional method that a large number of commodities in the operation record of the target user are simply used as recommended commodities, the commodities which are not purchased by the target user but are possibly interested can be recommended to the target user, so that the profit of merchants can be increased.
And 204, acquiring the file to be recommended and extracting the commodity to be recommended from the file to be recommended.
The document to be recommended may be a document to be recommended for recommending goods written by an operator employed by the merchant, and the document to be recommended may include one article to be recommended or a plurality of articles to be recommended, which is not limited herein.
And step 205, recommending the to-be-recommended file to the target user as a recommended file when the to-be-recommended commodity is the same as the interest commodity.
By adopting the steps to recommend the commodities to the user, the commodities can be recommended to the user accurately, and the commodities can be recommended without distinction, so that profits can be brought to merchants on the basis of keeping the stickiness of the user.
When the to-be-recommended case is processed, the to-be-recommended case comprises at least one commodity title, the commodity title and the commodity can have a corresponding relation, and the computer equipment stores a commodity title word segmentation word bank which comprises a plurality of commodity titles. The embodiment of the application provides an example of extracting a to-be-recommended commodity from a to-be-recommended file, and as shown in fig. 2, the steps can be 204-1 and 204-2.
And 204-1, segmenting words of the to-be-recommended case through the commodity title segmentation word library to obtain at least one commodity title of the to-be-recommended case.
In the embodiment of the application, the commodity appellation word library can be preset and comprises a plurality of commodity appellations, the commodity appellations correspond to a plurality of commodities which the merchants want to recommend, and the commodity appellation word library can be built by the user or obtained by adding the commodity appellations needing to be recommended to the existing word library.
And 204-2, taking the commodity corresponding to the commodity title of which the occurrence frequency exceeds a preset threshold value in at least one commodity title as the commodity to be recommended.
The merchant may recommend one or more commodities in the to-be-processed document, and except for the commodity that the merchant wants to recommend, other commodities may appear in the case of comparing the advantages of the recommended commodities, a threshold may be set for the number of occurrences of the commodity title, for example, ten times, that is, in the to-be-processed document, if the number of occurrences of the commodity title corresponding to a certain commodity exceeds ten times, the commodity may be considered as a mainly recommended commodity in the to-be-recommended document in which the commodity is located, and if the number of occurrences of the commodity title corresponding to a certain commodity does not exceed ten times after the word segmentation, the commodity may be considered as an additionally recommended commodity or a commodity subjected to reference comparison with the mainly recommended commodity. Through the mode of setting up the preset threshold value, compare and only take out the commodity that the number of times is the most now and regard as the reference in prior art, can more laminate actual conditions and acquire the commodity of treating recommending from treating the recommended case, and then improve the accuracy that the commodity was recommended.
On the basis of the foregoing, the commodity name may include a commodity name and/or a commodity short name and/or a commodity nickname, and the commodity name participle word bank may further include a corresponding relationship between the commodity name, the commodity short name and the commodity nickname. The embodiment of the application provides an example of obtaining at least one commodity title of a to-be-recommended case by segmenting words of the to-be-recommended case through a commodity title segmentation word bank, and the method can be realized through the following steps.
And performing word segmentation on the to-be-recommended case through a commodity name meaning word segmentation word bank to obtain at least one commodity name and/or commodity short name and/or commodity nickname.
The commodity names can comprise commodity names and/or commodity short names and/or commodity nicknames, such as the commodity to be recommended, "magical water", also called "SK 2", and also called "SK-II skin care essence lotion". Wherein, the SK-II skin care essence can be the trade name of the skin care essence, the SK2 can be the short name of the skin care essence, and the Shenxian water can be the nickname of the skin care essence. In the actual written case of the operator, the commodity name, the commodity short name and the commodity nickname may appear, and when a commodity appellation word library is constructed, the commodity name, the commodity short name and the commodity nickname of a commodity are all input into the appellation word library, so that the appearing words are used as participles in the word segmentation process, and the participles can be used when the appearance frequency of the commodity is calculated later.
On the basis of the foregoing, an example is provided in which a product corresponding to a product title whose occurrence frequency exceeds a preset threshold in at least one product title is used as a product to be recommended, and the method may be implemented through the following steps.
And according to the corresponding relation among the commodity name, the commodity short name and the commodity nickname, taking the commodity corresponding to the commodity name and/or the commodity short name and/or the commodity nickname with the occurrence frequency exceeding a preset threshold value as the commodity to be recommended.
On the basis of the above, as long as the product name, product short name and product nickname of the same product appear, the appearance times of the corresponding product are recorded. By adopting the mode to record the occurrence frequency of the commodities and further determine the commodities to be recommended, the situation that the literature is completely participled can be ensured, and meanwhile, the situation can be more suitable for actual operation (in the writing of the actual literature, official names of the commodities, namely commodity names, are generally not only used, but also commodity short names and commodity nicknames are used more in commodity recommendation and introduction).
Besides the aforementioned goods to be recommended are obtained by means of word segmentation, the document to be recommended may also include a link to the goods to be recommended; the embodiment of the application also provides an example of extracting the to-be-recommended goods from the to-be-recommended file, and as shown in fig. 3, the steps can be 204-3 and 204-4.
And step 204-3, extracting the links of the commodities to be recommended from the file to be recommended.
In the process of writing the document by the operator, the best condition is that the user can receive the commodity recommendation in the document after reading the document, and then purchase the commodity. The document to be recommended may be provided with a to-be-recommended product link, and the to-be-recommended product link may be in the form of a URI (Universal Resource Identifier, abbreviated as Universal Resource Identifier). In other embodiments of the embodiment of the present application, the to-be-recommended product link may also be in the form of a URL (Uniform Resource Locator).
And 204-4, analyzing the to-be-recommended commodity link to obtain the to-be-recommended commodity corresponding to the to-be-recommended commodity link.
In the embodiment of the present application, it is a mature technology in the art to specifically analyze the link of the to-be-recommended product, and details are not described herein. Through the scheme, the characteristic that the file for recommending the commodities generally comprises the link of the commodities to be recommended in the actual situation is utilized, and the commodities to be recommended can be accurately obtained.
In addition to the foregoing processing flow performed on the article, an example of obtaining interested goods of the target user through collaborative filtering calculation according to the operation data of the target user on the target shopping platform and the operation data of each other user on the target shopping platform is provided in the embodiment of the present application, and please refer to fig. 4, which can be implemented through the following steps.
And 203-1, calculating to obtain the user similarity of the target user and each other user according to the operation data of the target user on the target shopping platform and the operation data of each other user on the target shopping platform.
In the embodiment of the present application, as described above, the interested goods of the target user may be determined by referring to other users having operation data on the same target shopping platform as the target user, specifically, the user similarity between the target user and other users on the target shopping platform may be calculated first to determine other users having high similarity between the operation data and the user, and specifically, a formula may be used:
calculating the similarity between the target user and other users, wherein N (u) is each commodity operated by the target user u in the target platform, | N (u) | is the operation frequency of the target user u for each commodity, N (v) is each commodity operated by other users v in the target platform, | N (v) | is the operation frequency of other users v for each commodity, and W (v) | is the operation frequency of other users v for each commodityuvThe similarity between the target user u and the other users v. By adopting the cosine similarity formula, the similarity between the target user u and other users v can be accurately obtained.
And step 203-2, determining a reference user with high similarity to the target user from the plurality of other users according to the user similarity of the target user and each other user.
A similarity threshold may be set, for example, 90%, and users whose similarity between the operation data of the target operation platform and the operation data of the target user exceeds 90% may be used as reference users to determine interesting commodities of the target user. In yet another implementation manner of the embodiment of the present application, the similarity threshold and the preset number of users may be set at the same time for consideration, for example, after ranking other users from high to low according to the similarity with the target user, the first 5 other users are taken, and then it is determined whether the user similarity between the 5 users and the target user exceeds 90%, and the excess is retained and not removed. By adopting the method to select the reference users, reliable reference data for determining the interest commodities of the target users can be selected.
And 203-3, calculating to obtain the interested commodity of the target user according to the operation data of the reference user on the target shopping platform and the operation data of the target user on the target shopping platform.
After the reference user is determined, calculation can be performed according to the operation data of the reference user on the target shopping platform and the operation data of the target user on the target shopping platform to obtain the interested commodity of the target user.
On the basis of the above, the operation data of the reference user on the target shopping platform can comprise the reference commodity. The embodiment of the application provides an example of calculating the interested commodity of the target user according to the operation data of the reference user on the target shopping platform and the operation data of the target user on the target shopping platform, and as shown in fig. 5, the following steps can be implemented.
And step 203-3-1, calculating the grade of the reference commodity included in the operation data of the target shopping platform by the reference user through an entropy weight method according to the operation data of the reference user on the target shopping platform.
On the basis, the score of the reference user on the reference commodity included in the operation data of the target shopping platform can be calculated and obtained firstly, so as to prepare for subsequently determining the interested commodity of the target user, specifically, in the embodiment of the application, the reference user can click, purchase, add to a shopping cart and collect the commodity, and a decision matrix can be constructed:
wherein d is1To click, d2For purchase, d3To add to a shopping cart, d4For collection, inFor the purpose of reference to the commercial product,
for reference user to goods inD is performed num timeskOperations, in the present application example, k is 1, 2, 3, 4, and in other implementations of the present application example, other operations, such as repurchase, may also be included.
After the decision matrix X is obtained, it may be normalized and each sample value calculated (e.g., for example)
) In this operation (d)1) The following specific gravity can be calculated according to the formula:
calculated, where i 1., n, j 1., m, PijIs the j index (i.e. d)1，d2，d3，d4) The lower ith sample value (e.g.
) The specific gravity of the index. On the basis, the entropy value of the j index can be further calculated, and can be represented by the formula:
calculating, wherein k is 1/ln (n) and satisfies ejIs more than or equal to 0. Then, the information entropy redundancy can be further calculated, and the information entropy redundancy can be obtained through a formula:
dj＝1-ej，j＝1，...，m
wherein d isjThe information entropy redundancy of the j index. The following equation may then be followed:
and calculating to obtain the weight of the j index. After the entropy weights are determined, it is possible to calculate the entropy weights according to the formula:
calculating the grade of the reference user for the reference commodity included in the operation data of the target shopping platform, wherein siFor the score of the reference commodity i, k is the operation type of the reference user on the reference commodity i on the target shopping platform, and wk is the weight of the operation type k (i.e. the aforementioned wj)，numkThe number of executions of operation type k.
And step 203-3-2, calculating to obtain the interest value of the target user on the reference commodity through a preset formula according to the score of the reference commodity and the user similarity between the reference user and the target user.
Based on the above, we can get the rating of each reference user for the reference goods included in the operation data of the target shopping platform, and therefore can be based on the formula:
wherein p (u, i) is the score of the target user u on the reference commodity i, and wuvIs the similarity of the target user u and the reference user v, rviTo reference the rating of the user v for the reference item i, and v ∈ S (u,K) ∩ n (i) indicates that the reference user v is the first K other users which are obtained from the other users having operation data coinciding with the target user u according to the preset similarity threshold value and serve as the reference user as described above.
And 203-3-3, when the interest value of the target user on the reference commodity exceeds a preset interest threshold value, taking the reference commodity as the interest commodity of the target user.
It should be appreciated that through the use of the entropy weighting method and the user collaborative filtering method, the commodities which may be of interest to the user but have not yet been purchased can be recommended to the user, thereby increasing the profit of the merchant. For example, because the commodity a has a higher similarity, a given examination user still has a higher score for the commodity B, and the target user does not have any operation on the commodity B, through the above-mentioned process, the score of the commodity B obtained through the final calculation for the target user is also higher, and the product can be used as an interesting commodity of the target user, so that the matched file can be pushed to the target user.
An embodiment of the present application further provides a product recommendation apparatus 110, which is applied to a computer device, where the computer device stores operation data of a plurality of users on a target shopping platform, as shown in fig. 6, the apparatus includes:
an obtaining module 1101, configured to obtain operation data of a target user on a target shopping platform; and acquiring operation data of a plurality of other users on the target shopping platform, wherein the other users are users except the target user in the plurality of users.
The calculation module 1102 is configured to obtain the interested goods of the target user through collaborative filtering calculation according to the operation data of the target user on the target shopping platform and the operation data of each of the other users on the target shopping platform.
The recommendation module 1103 is used for acquiring the document to be recommended and extracting the commodity to be recommended from the document to be recommended; and when the to-be-recommended commodities are the same as the interested commodities, recommending the to-be-recommended file to the target user as a recommended file.
Further, the to-be-recommended case comprises at least one commodity title, the commodity title has a corresponding relation with a commodity, the computer device stores a commodity title word segmentation word bank, and the commodity title word segmentation word bank comprises a plurality of commodity titles;
the recommendation module 1103 is specifically configured to:
segmenting words of the to-be-recommended case through the commodity title segmentation word bank to obtain at least one commodity title of the to-be-recommended case; and taking the commodity corresponding to the commodity title of which the occurrence frequency exceeds a preset threshold value in the at least one commodity title as the commodity to be recommended.
Further, the commodity name comprises a commodity name and/or a commodity short name and/or a commodity nickname, and the commodity name participle word bank also comprises a corresponding relation among the commodity name, the commodity short name and the commodity nickname; the recommending module 1103 is further specifically configured to:
and performing word segmentation on the to-be-recommended case through the commodity title word segmentation word bank to obtain at least one commodity name and/or commodity short name and/or commodity nickname.
The recommending module 1103 is further specifically configured to:
and according to the corresponding relation among the commodity names, the short commodity names and the nicknames of the commodities, the commodities corresponding to the commodity names and/or the short commodity names and/or the nicknames of the commodities, the occurrence times of which exceed a preset threshold value, are taken as the commodities to be recommended.
Further, the document to be recommended also comprises a commodity link to be recommended;
the recommendation module 1103 is specifically configured to:
extracting the commodity link to be recommended from the file to be recommended; and analyzing the commodity link to be recommended to obtain the commodity to be recommended corresponding to the commodity link to be recommended.
Further, the calculation module 1102 is specifically configured to:
calculating the user similarity of the target user and each other user according to the operation data of the target user on the target shopping platform and the operation data of each other user on the target shopping platform; according to the user similarity between the target user and each of the other users, determining a reference user with high similarity to the target user from the other users; and calculating to obtain the interested commodity of the target user according to the operation data of the reference user on the target shopping platform and the operation data of the target user on the target shopping platform.
Further, the operation data of the reference user on the target shopping platform comprises reference commodities;
the calculating module 1102 is further specifically configured to:
calculating the grade of the reference user on the reference commodity in the operation data of the target shopping platform through an entropy weight method according to the operation data of the reference user on the target shopping platform; calculating to obtain an interest value of the target user to the reference commodity through a preset formula according to the score of the reference commodity and the user similarity of the reference user and the target user; and when the interest value of the target user to the reference commodity exceeds a preset interest threshold value, taking the reference commodity as the interest commodity of the target user.
Further, the score of the reference commodity is calculated by the following formula:
wherein s isiThe grade of the reference commodity i is determined, k is the operation type of the reference user on the reference commodity i on the target shopping platform, wkIs the weight of the operation type k, numkIs the number of executions of the operation type k.
In an embodiment of the present application, the computer device 100 includes a processor and a non-volatile memory storing computer instructions, and when the computer instructions are executed by the processor, the computer device 100 executes the aforementioned product recommendation method. As shown in fig. 7, fig. 7 is a block diagram of a computer device 100 according to an embodiment of the present disclosure. The computer device 100 includes a commodity recommending apparatus 110, a memory 111, a processor 112, and a communication unit 113.
The memory 111, the processor 112 and the communication unit 113 are electrically connected to each other directly or indirectly to realize data transmission or interaction. For example, the components may be electrically connected to each other via one or more communication buses or signal lines. The commodity recommending apparatus 110 includes at least one software function module which can be stored in the memory 111 in the form of software or firmware (firmware) or is fixed in an Operating System (OS) of the computer device 100. The processor 112 is used for executing executable modules stored in the memory 111, such as software functional modules and computer programs included in the product recommendation device 110.
The Memory 111 may be, but is not limited to, a Random Access Memory (RAM), a Read Only Memory (ROM), a Programmable Read-Only Memory (PROM), an Erasable Read-Only Memory (EPROM), an electrically Erasable Read-Only Memory (EEPROM), and the like.
An embodiment of the present application provides a readable storage medium, where the readable storage medium includes a computer program, and when the computer program runs, the computer device where the readable storage medium is located is controlled to execute the aforementioned commodity recommendation method.
In summary, the embodiment of the present application provides a method, an apparatus, a computer device and a readable storage medium for recommending a commodity, by acquiring operation data of a target user on a target shopping platform, and referring to the operation data of the target user on the target shopping platform by a plurality of other users; then obtaining interested commodities of the target user through collaborative filtering calculation according to the operation data of the target user on the target shopping platform and the operation data of each other user on the target shopping platform; then obtaining a document to be recommended and extracting a commodity to be recommended from the document to be recommended; when the goods to be recommended are the same as the interested goods, the documents to be recommended are recommended to the target user as the recommended documents, so that the appropriate goods can be accurately recommended to the target user, the target user of the recommended goods does not pay attention to the recommended goods, the recommended goods is attractive, the sales volume of the merchant is increased, the profit is improved, and the user stickiness is further increased.
The above description is only for the specific embodiments of the present application, but the scope of the present application is not limited thereto, and any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope of the present application should be covered within the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
Claims (10)
1. A commodity recommendation method is applied to a computer device, wherein the computer device stores operation data of a plurality of users on a target shopping platform, and the method comprises the following steps:
acquiring operation data of a target user on the target shopping platform;
acquiring operation data of a plurality of other users in the target shopping platform, wherein the other users are users except the target user in the plurality of users;
obtaining interested commodities of the target user through collaborative filtering calculation according to the operation data of the target user on the target shopping platform and the operation data of each other user on the target shopping platform;
acquiring a to-be-recommended file and extracting a to-be-recommended commodity from the to-be-recommended file;
and when the to-be-recommended commodity is the same as the interested commodity, recommending the to-be-recommended file to the target user as a recommended file.
2. The method according to claim 1, wherein the document to be recommended includes at least one commodity title, the commodity title has a corresponding relationship with a commodity, the computer device stores a commodity title participle thesaurus, and the commodity title participle thesaurus includes a plurality of commodity titles;
the step of extracting the goods to be recommended from the file to be recommended comprises the following steps:
segmenting words of the to-be-recommended case through the commodity title segmentation word bank to obtain at least one commodity title of the to-be-recommended case;
and taking the commodity corresponding to the commodity title of which the occurrence frequency exceeds a preset threshold value in the at least one commodity title as the commodity to be recommended.
3. The method of claim 2, wherein the commodity names comprise commodity names and/or short commodity names and/or nicknames, and the commodity name participle thesaurus further comprises correspondence among the commodity names, short commodity names and nicknames;
the step of performing word segmentation on the to-be-recommended case through the commodity title word segmentation word bank to obtain at least one commodity title of the to-be-recommended case comprises the following steps of:
performing word segmentation on the to-be-recommended case through the commodity title word segmentation word bank to obtain at least one commodity name and/or commodity short name and/or commodity nickname;
the step of taking the commodity corresponding to the commodity title of which the occurrence frequency exceeds a preset threshold value in the at least one commodity title as the commodity to be recommended comprises the following steps:
and according to the corresponding relation among the commodity names, the short commodity names and the nicknames of the commodities, the commodities corresponding to the commodity names and/or the short commodity names and/or the nicknames of the commodities, the occurrence times of which exceed a preset threshold value, are taken as the commodities to be recommended.
4. The method according to claim 1, wherein the document to be recommended further comprises a link to a commodity to be recommended;
the step of extracting the goods to be recommended from the file to be recommended comprises the following steps:
extracting the commodity link to be recommended from the file to be recommended;
and analyzing the commodity link to be recommended to obtain the commodity to be recommended corresponding to the commodity link to be recommended.
5. The method of claim 1, wherein the step of obtaining the target user's interest goods through collaborative filtering calculation according to the operation data of the target user on the target shopping platform and the operation data of each of the other users on the target shopping platform comprises:
calculating the user similarity of the target user and each other user according to the operation data of the target user on the target shopping platform and the operation data of each other user on the target shopping platform;
according to the user similarity between the target user and each of the other users, determining a reference user with high similarity to the target user from the other users;
and calculating to obtain the interested commodity of the target user according to the operation data of the reference user on the target shopping platform and the operation data of the target user on the target shopping platform.
6. The method of claim 5, wherein the operational data of the reference user at the target shopping platform includes reference merchandise;
the step of calculating the interest goods of the target user according to the operation data of the reference user on the target shopping platform and the operation data of the target user on the target shopping platform comprises the following steps:
calculating the grade of the reference user on the reference commodity in the operation data of the target shopping platform through an entropy weight method according to the operation data of the reference user on the target shopping platform;
calculating to obtain an interest value of the target user to the reference commodity through a preset formula according to the score of the reference commodity and the user similarity of the reference user and the target user;
and when the interest value of the target user to the reference commodity exceeds a preset interest threshold value, taking the reference commodity as the interest commodity of the target user.
7. The method of claim 6, wherein the score of the reference good is calculated by the formula:
wherein s isiThe grade of the reference commodity i is determined, k is the operation type of the reference user on the reference commodity i on the target shopping platform, wkIs the weight of the operation type k, numkIs the number of executions of the operation type k.
8. An article recommendation apparatus applied to a computer device storing operation data of a plurality of users on a target shopping platform, the apparatus comprising:
the acquisition module is used for acquiring the operation data of the target user on the target shopping platform; acquiring operation data of a plurality of other users in the target shopping platform, wherein the other users are users except the target user in the plurality of users;
the calculation module is used for obtaining interested commodities of the target user through collaborative filtering calculation according to the operation data of the target user on the target shopping platform and the operation data of each other user on the target shopping platform;
the recommendation module is used for acquiring a to-be-recommended file and extracting a to-be-recommended commodity from the to-be-recommended file; and when the to-be-recommended commodity is the same as the interested commodity, recommending the to-be-recommended file to the target user as a recommended file.
9. A computer device comprising a processor and a non-volatile memory storing computer instructions that, when executed by the processor, perform the item recommendation method of any of claims 1-7.
10. A readable storage medium, characterized in that the readable storage medium comprises a computer program, and the computer program controls a computer device in which the readable storage medium is executed to execute the commodity recommendation method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010550930.XA CN111681086A (en) | 2020-06-16 | 2020-06-16 | Commodity recommendation method and device, computer equipment and readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010550930.XA CN111681086A (en) | 2020-06-16 | 2020-06-16 | Commodity recommendation method and device, computer equipment and readable storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111681086A true CN111681086A (en) | 2020-09-18 |
Family
ID=72436424
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010550930.XA Pending CN111681086A (en) | 2020-06-16 | 2020-06-16 | Commodity recommendation method and device, computer equipment and readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111681086A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113744020A (en) * | 2021-01-15 | 2021-12-03 | 北京沃东天骏信息技术有限公司 | Commodity file processing method and device, electronic equipment and storage medium |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150379609A1 (en) * | 2014-06-30 | 2015-12-31 | Kobo Incorporated | Generating recommendations for unfamiliar users by utilizing social side information |
| CN108304533A (en) * | 2018-01-29 | 2018-07-20 | 上海名轩软件科技有限公司 | Keyword recommendation method and equipment |
| CN109087177A (en) * | 2018-08-21 | 2018-12-25 | 北京京东尚科信息技术有限公司 | To the method, apparatus and computer readable storage medium of target user's Recommendations |
| CN109597973A (en) * | 2017-09-30 | 2019-04-09 | 阿里巴巴集团控股有限公司 | A kind of recommendation, generation method and the device of official documents and correspondence information |
| CN110674620A (en) * | 2019-09-20 | 2020-01-10 | 阿里巴巴（中国）有限公司 | Target file generation method, device, medium and electronic equipment |
| CN110706055A (en) * | 2019-08-22 | 2020-01-17 | 平安科技（深圳）有限公司 | Commodity information pushing method and device, storage medium and computer equipment |
| CN110956530A (en) * | 2019-11-26 | 2020-04-03 | 上海风秩科技有限公司 | Recommendation method and device, electronic equipment and computer-readable storage medium |
| CN111127152A (en) * | 2019-12-23 | 2020-05-08 | 深圳市赛维网络科技有限公司 | Commodity recommendation method, device and equipment based on user preference prediction and readable medium |
-
2020
- 2020-06-16 CN CN202010550930.XA patent/CN111681086A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150379609A1 (en) * | 2014-06-30 | 2015-12-31 | Kobo Incorporated | Generating recommendations for unfamiliar users by utilizing social side information |
| CN109597973A (en) * | 2017-09-30 | 2019-04-09 | 阿里巴巴集团控股有限公司 | A kind of recommendation, generation method and the device of official documents and correspondence information |
| CN108304533A (en) * | 2018-01-29 | 2018-07-20 | 上海名轩软件科技有限公司 | Keyword recommendation method and equipment |
| CN109087177A (en) * | 2018-08-21 | 2018-12-25 | 北京京东尚科信息技术有限公司 | To the method, apparatus and computer readable storage medium of target user's Recommendations |
| CN110706055A (en) * | 2019-08-22 | 2020-01-17 | 平安科技（深圳）有限公司 | Commodity information pushing method and device, storage medium and computer equipment |
| CN110674620A (en) * | 2019-09-20 | 2020-01-10 | 阿里巴巴（中国）有限公司 | Target file generation method, device, medium and electronic equipment |
| CN110956530A (en) * | 2019-11-26 | 2020-04-03 | 上海风秩科技有限公司 | Recommendation method and device, electronic equipment and computer-readable storage medium |
| CN111127152A (en) * | 2019-12-23 | 2020-05-08 | 深圳市赛维网络科技有限公司 | Commodity recommendation method, device and equipment based on user preference prediction and readable medium |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113744020A (en) * | 2021-01-15 | 2021-12-03 | 北京沃东天骏信息技术有限公司 | Commodity file processing method and device, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109191240B (en) | Method and device for recommending commodities | |
| CN109697629B (en) | Product data pushing method and device, storage medium and computer equipment | |
| CN108805598B (en) | Similarity information determination method, server and computer-readable storage medium | |
| CN108921398B (en) | Shop quality evaluation method and device | |
| CN111325609A (en) | Commodity recommendation list determining method and device, electronic equipment and storage medium | |
| US20220180379A1 (en) | Transaction-based information processing system, method, and article | |
| US20190080352A1 (en) | Segment Extension Based on Lookalike Selection | |
| CN111861605B (en) | Service object recommendation method | |
| EP3779835A1 (en) | Device, method and program for analyzing customer attribute information | |
| CN113516496B (en) | Advertisement conversion rate estimation model construction method, device, equipment and medium thereof | |
| KR20150121281A (en) | A method for recommending an item based on a comstomer's preference to the item and a computer readable recording medium to store instructions to perform the method | |
| CN112765230B (en) | Payment big data analysis method and big data analysis system based on internet finance | |
| CN110674391A (en) | Product data pushing method and system based on big data and computer equipment | |
| CN111582932A (en) | Inter-scene information pushing method and device, computer equipment and storage medium | |
| CN111754278A (en) | Article recommendation method and device, computer storage medium and electronic equipment | |
| CN111429214A (en) | Transaction data-based buyer and seller matching method and device | |
| CN111681086A (en) | Commodity recommendation method and device, computer equipment and readable storage medium | |
| CN113837843A (en) | Product recommendation method, device, medium and electronic equipment | |
| Margalina et al. | Similarities in factors affecting online shopping intention in Ecuador and Peru: a multigroup analysis | |
| CN111311351A (en) | File recommendation method and device, computer equipment and readable storage medium | |
| CN115660733A (en) | Sales prediction system and method based on artificial intelligence | |
| CN112015970A (en) | Product recommendation method, related equipment and computer storage medium | |
| CN113077292B (en) | User classification method and device, storage medium and electronic equipment | |
| CN111639989B (en) | Commodity recommendation method and readable storage medium | |
| CN112069388B (en) | Entity recommendation method, system, computer device and computer readable storage 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 | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20200918 |
|
| WD01 | Invention patent application deemed withdrawn after publication |