CN114090880A - Method and device for commodity recommendation, electronic equipment and storage medium - Google Patents

Method and device for commodity recommendation, electronic equipment and storage medium Download PDF

Info

Publication number
CN114090880A
CN114090880A CN202111327653.7A CN202111327653A CN114090880A CN 114090880 A CN114090880 A CN 114090880A CN 202111327653 A CN202111327653 A CN 202111327653A CN 114090880 A CN114090880 A CN 114090880A
Authority
CN
China
Prior art keywords
commodity
commodities
data
recommended
attribute
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
Application number
CN202111327653.7A
Other languages
Chinese (zh)
Inventor
李犇
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beijing Mininglamp Software System Co ltd
Original Assignee
Beijing Mininglamp Software System Co ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Beijing Mininglamp Software System Co ltd filed Critical Beijing Mininglamp Software System Co ltd
Priority to CN202111327653.7A priority Critical patent/CN114090880A/en
Publication of CN114090880A publication Critical patent/CN114090880A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/95—Retrieval from the web
    • G06F16/953—Querying, e.g. by the use of web search engines
    • G06F16/9535—Search customisation based on user profiles and personalisation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/95—Retrieval from the web
    • G06F16/953—Querying, e.g. by the use of web search engines
    • G06F16/9536—Search customisation based on social or collaborative filtering
    • 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]
    • 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
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/01—Social networking

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Theoretical Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Accounting & Taxation (AREA)
  • Finance (AREA)
  • Marketing (AREA)
  • General Business, Economics & Management (AREA)
  • Strategic Management (AREA)
  • Economics (AREA)
  • Development Economics (AREA)
  • General Engineering & Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Computing Systems (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Human Resources & Organizations (AREA)
  • Primary Health Care (AREA)
  • Tourism & Hospitality (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

The application relates to the technical field of computers, and discloses a method for recommending commodities, which comprises the following steps: the method comprises the steps of obtaining comment data of a plurality of seed users on different commodities, attribute data corresponding to the commodities and behavior data corresponding to the seed users; determining commodities to be recommended according to the attribute data and the comment data, and carrying out user clustering according to the behavior data to obtain a user cluster; and recommending the commodity to be recommended to other users outside the seed user corresponding to the commodity to be recommended in the user cluster. Therefore, the commodities to be recommended are determined through the collected comment data and attribute data, the user cluster is obtained through user clustering according to the collected behavior data, the commodities to be recommended are recommended to the users in the user cluster, and commodity recommendation of the crowd related to the users can be achieved. The application also discloses a device for recommending commodities, electronic equipment and a storage medium.

Description

Method and device for commodity recommendation, electronic equipment and storage medium
Technical Field
The present application relates to the field of computer technologies, and in particular, to a method and an apparatus for recommending a commodity, an electronic device, and a storage medium.
Background
With the development and popularization of internet technology and electronic commerce, more and more internet users emerge, and people are gradually used to perform online shopping on an internet shopping platform. The internet shopping platform usually predicts shopping preferences or shopping demands of a user according to shopping behaviors, browsing behaviors, collecting behaviors and the like of the user, so as to recommend commodities to the user.
In the process of implementing the embodiments of the present disclosure, it is found that at least the following problems exist in the related art:
in the prior art, when recommending commodities to a plurality of users, the commodities of the same type as the favorite commodities of each user are generally recommended to each user by determining the favorite commodities of each user, so that the commodity recommendation to the crowd related to the user is difficult to realize.
Disclosure of Invention
The following presents a simplified summary in order to provide a basic understanding of some aspects of the disclosed embodiments. This summary is not an extensive overview nor is intended to identify key/critical elements or to delineate the scope of such embodiments but rather as a prelude to the more detailed description that is presented later.
The embodiment of the disclosure provides a method and a device for commodity recommendation, electronic equipment and a storage medium, so that commodity recommendation of people related to a user can be realized.
In some embodiments, the method for merchandise recommendation includes: the method comprises the steps of obtaining comment data of a plurality of seed users on different commodities, attribute data corresponding to the commodities and behavior data corresponding to the seed users; determining a commodity to be recommended according to the attribute data and the comment data, and performing user clustering according to the behavior data to obtain a user cluster; recommending the commodity to be recommended to other users outside the seed user corresponding to the commodity to be recommended in the user cluster.
In some embodiments, the apparatus for merchandise recommendation includes: the acquisition module is configured to acquire comment data of a plurality of seed users on different commodities, attribute data corresponding to the commodities and behavior data corresponding to the seed users; the determining module is configured to determine commodities to be recommended according to the attribute data and the comment data, and perform user clustering according to the behavior data to obtain a user cluster; and the recommending module is configured to recommend the commodity to be recommended to other users outside the seed user corresponding to the commodity to be recommended in the user cluster.
In some embodiments, the electronic device comprises a processor and a memory storing program instructions, the processor being configured to, when executing the program instructions, perform the method for merchandise recommendation as described above.
In some embodiments, the storage medium stores program instructions that, when executed, perform the method for merchandise recommendation described above.
The method and the device for recommending commodities, the electronic equipment and the storage medium provided by the embodiment of the disclosure can realize the following technical effects: the method comprises the steps that comment data of a plurality of seed users on different commodities, attribute data corresponding to the commodities and behavior data corresponding to the seed users are obtained; determining commodities to be recommended according to the attribute data and the comment data, and carrying out user clustering according to the behavior data to obtain a user cluster; and recommending the commodity to be recommended to other users outside the seed user corresponding to the commodity to be recommended in the user cluster. Therefore, the commodities to be recommended are determined through the collected comment data and attribute data, the user cluster is obtained through user clustering according to the collected behavior data, the commodities to be recommended are recommended to the users in the user cluster, and commodity recommendation of the crowd related to the users can be achieved.
The foregoing general description and the following description are exemplary and explanatory only and are not restrictive of the application.
Drawings
One or more embodiments are illustrated by way of example in the accompanying drawings, which correspond to the accompanying drawings and not in limitation thereof, in which elements having the same reference numeral designations are shown as like elements and not in limitation thereof, and wherein:
FIG. 1 is a schematic diagram of a method for merchandise recommendation provided by embodiments of the present disclosure;
FIG. 2 is a schematic diagram of another method for merchandise recommendation provided by embodiments of the present disclosure;
FIG. 3 is a schematic diagram of a method for determining an item to be recommended according to an embodiment of the disclosure;
FIG. 4 is a schematic diagram of a method for determining a user cluster according to an embodiment of the present disclosure;
FIG. 5 is a schematic diagram of an apparatus for merchandise recommendation provided by embodiments of the present disclosure;
fig. 6 is a schematic diagram of an electronic device provided by an embodiment of the disclosure.
Detailed Description
So that the manner in which the features and elements of the disclosed embodiments can be understood in detail, a more particular description of the disclosed embodiments, briefly summarized above, may be had by reference to the embodiments, some of which are illustrated in the appended drawings. In the following description of the technology, for purposes of explanation, numerous details are set forth in order to provide a thorough understanding of the disclosed embodiments. However, one or more embodiments may be practiced without these details. In other instances, well-known structures and devices may be shown in simplified form in order to simplify the drawing.
The terms "first," "second," and the like in the description and in the claims, and the above-described drawings of embodiments of the present disclosure, are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It should be understood that the data so used may be interchanged under appropriate circumstances such that embodiments of the present disclosure described herein may be made. Furthermore, the terms "comprising" and "having," as well as any variations thereof, are intended to cover non-exclusive inclusions.
The term "plurality" means two or more unless otherwise specified.
In the embodiment of the present disclosure, the character "/" indicates that the preceding and following objects are in an or relationship. For example, A/B represents: a or B.
The term "and/or" is an associative relationship that describes objects, meaning that three relationships may exist. For example, a and/or B, represents: a or B, or A and B.
The term "correspond" may refer to an association or binding relationship, and a corresponds to B refers to an association or binding relationship between a and B.
With reference to fig. 1, an embodiment of the present disclosure provides a method for recommending a commodity, including:
step S101, obtaining comment data of a plurality of seed users to different commodities, attribute data corresponding to the commodities and behavior data corresponding to the seed users.
And S102, determining the commodity to be recommended according to the attribute data and the comment data, and carrying out user clustering according to the behavior data to obtain a user cluster.
And step S103, recommending the commodity to be recommended to other users outside the seed user corresponding to the commodity to be recommended in the user cluster.
By adopting the method for recommending commodities provided by the embodiment of the disclosure, the comment data of a plurality of seed users to different commodities, the attribute data corresponding to the commodities and the behavior data corresponding to the seed users are obtained; determining commodities to be recommended according to the attribute data and the comment data, and carrying out user clustering according to the behavior data to obtain a user cluster; and recommending the commodity to be recommended to other users outside the seed user corresponding to the commodity to be recommended in the user cluster. Therefore, the commodities to be recommended are determined through the collected comment data and attribute data, the user cluster is obtained through user clustering according to the collected behavior data, the commodities to be recommended are recommended to the users in the user cluster, and commodity recommendation of the crowd related to the users can be achieved.
Optionally, the attribute data of the commodity includes commodity information such as brand, category, composition, appearance, efficacy, and material.
Optionally, the behavior data corresponding to the user includes behaviors of browsing, searching, and purchasing.
In some embodiments, comment data of the commodities, attribute data corresponding to the commodities and behavior data corresponding to various sub-users are collected from the e-commerce platform. Therefore, by acquiring the comment data of the commodity, the comment data can represent the preference of the user to the commodity, the requirements and the attention points of the user on different aspects of the commodity, and the commodity to be recommended which meets the preference of the user can be more conveniently determined.
Optionally, determining the to-be-recommended item according to the attribute data and the comment data includes: acquiring attribute embedded vectors corresponding to the commodities according to the attribute data, and acquiring comment embedded vectors corresponding to the commodities according to the comment data; fusing and splicing the attribute embedded vectors and the comment embedded vectors to obtain commodity embedded vectors corresponding to commodities; and determining the commodities to be recommended according to the commodity embedded vectors.
Optionally, obtaining an attribute embedded vector corresponding to the commodity according to the attribute data includes: acquiring a first feature vector corresponding to attribute data; and calculating by using the first feature vector according to a preset first algorithm to obtain an attribute embedded vector corresponding to the commodity.
Optionally, attributes of the product, such as a brand, a category, a component, an appearance, a power, or a material, are extracted from the attribute data by using a regular expression or an NER (Named Entity Recognition) algorithm model.
Optionally, obtaining a first feature vector corresponding to the attribute data includes: constructing a commodity attribute map according to the attribute data; and determining a first feature vector corresponding to the attribute data according to the commodity attribute map.
Optionally, constructing a product attribute map corresponding to the product according to the attribute data includes: determining the name of the commodity as a first root node, and determining the attribute of the commodity as a first leaf node corresponding to the first root node; and constructing a commodity attribute map according to the first root node and the first leaf node.
In some embodiments, a product attribute map corresponding to each product is obtained according to each attribute data.
Optionally, determining a first feature vector corresponding to the attribute data according to the commodity attribute map includes: initializing a first leaf node of a commodity attribute map according to a BERT (Bidirectional Encoder Representation from converters) pre-training model to obtain a first feature vector.
In some embodiments, each attribute in the commodity attribute map is determined as an input feature of the BERT pre-trained model, such as: "[ CLS ] attribute [ SEP ]"; where [ CLS ] is the beginning marker of the input feature and [ SEP ] is the separation marker and the end marker of the input feature. Inputting the attribute [ SEP ] of [ CLS ] into a BERT pre-training model to obtain a first feature vector. Therefore, by constructing the commodity attribute map, the commodity attributes concerned by the user can be better displayed, and meanwhile, the commodities to be recommended, which accord with the user preferences, can be more conveniently determined.
Optionally, calculating by using the first feature vector according to a preset first algorithm to obtain an attribute embedded vector corresponding to the commodity, including: by calculation of
Figure BDA0003347505000000051
Obtaining an attribute embedded vector corresponding to the commodity; wherein h isv1Embedding a vector for the attribute of the first root node v1, namely embedding a vector for the attribute corresponding to the commodity; n is a radical ofv1Is the neighborhood of the first root node v 1; h isu1A first eigenvector corresponding to the first leaf node u1 in the neighborhood of the first root node v 1.
Optionally, when the first leaf node is a terminal node, that is, when there is no child node corresponding to the first leaf node, calculating by using the first feature vector according to a preset first algorithm to obtain an attribute embedded vector corresponding to the commodity.
Optionally, when the first leaf node is not a terminal node, that is, when there is a child node corresponding to the first leaf node, obtaining a fifth feature vector corresponding to each child node, and performing calculation by using the fifth feature vector according to a preset fourth algorithm to obtain a first feature vector corresponding to the first leaf node; and calculating by utilizing the first feature vector according to a first algorithm to obtain the embedded vector of the corresponding attribute of the commodity. Therefore, under the condition that the commodity attribute map is multilayer, the feature vectors of the nodes of each layer can be respectively aggregated through different preset algorithms, and the attribute embedded vector corresponding to the commodity is obtained.
Optionally, the obtaining a fifth feature vector corresponding to each child node includes: and initializing each child node according to the BERT pre-training model to obtain a fifth feature vector.
Optionally according to presetsThe fourth algorithm calculates by using the fifth feature vector to obtain the first feature vector corresponding to the first leaf node, and includes: by calculation of
Figure BDA0003347505000000061
Obtaining a first feature vector corresponding to a first leaf node; wherein h isu1A first feature vector corresponding to a first leaf node u 1; n is a radical ofu1Neighborhood of the first leaf node u 1; h ismA fifth eigenvector corresponding to child node m in the neighborhood of the first leaf node u 1.
As shown in fig. 2, an embodiment of the present disclosure provides a method for recommending a commodity, including:
step S201, obtaining comment data of a plurality of seed users on different commodities, attribute data corresponding to each commodity, and behavior data corresponding to each seed user.
Step S202, attribute embedded vectors corresponding to the commodities are obtained according to the attribute data, and comment embedded vectors corresponding to the commodities are obtained according to the comment data.
And step S203, fusing and splicing the attribute embedded vectors and the comment embedded vectors to obtain commodity embedded vectors corresponding to commodities.
And S204, determining the commodities to be recommended according to the commodity embedding vectors, and carrying out user clustering according to the behavior data to obtain a user cluster.
And step S205, recommending the commodity to be recommended to other users outside the seed user corresponding to the commodity to be recommended in the user cluster.
By adopting the method for recommending commodities provided by the embodiment of the disclosure, the attribute embedded vector corresponding to the commodity is obtained through the collected attribute data, and the comment embedded vector corresponding to the commodity is obtained according to the comment data; fusing and splicing the attribute embedded vector corresponding to each commodity and the comment embedded vector to obtain a commodity embedded vector corresponding to each commodity; the commodities to be recommended which are more suitable for the user preference can be determined according to the commodity embedding vector, the user cluster is obtained by carrying out user clustering according to the collected behavior data, the commodities to be recommended are recommended to the users in the user cluster, and commodity recommendation of the crowd related to the users can be achieved.
Optionally, obtaining a comment embedding vector corresponding to the commodity according to the comment data includes: acquiring a second feature vector corresponding to the comment data, and acquiring an emotion score corresponding to the commodity according to the comment data; fusing and splicing the second feature vector and the emotion score to obtain a third feature vector; and calculating by using the third feature vector according to a preset second algorithm to obtain a comment embedded vector corresponding to the commodity.
Optionally, obtaining an emotion score corresponding to the commodity according to the comment data includes: obtaining the emotion type corresponding to the attribute of the commodity according to the comment data; the emotional type includes positive, negative or neutral; counting the emotion types corresponding to the attributes of the commodities to obtain the probability that the emotion types are positive; and determining the probability that the emotion type is positive as the emotion score corresponding to the commodity.
Optionally, obtaining a second feature vector corresponding to the comment data includes: building a commodity comment map according to each comment data; and determining a second feature vector corresponding to the comment data according to the commodity comment map.
Optionally, constructing a product review graph corresponding to the product according to the review data includes: determining the name of the commodity as a second root node, and determining the probability that the emotion type corresponding to the attribute of the commodity is positive as a second leaf node corresponding to the second root node; and constructing a commodity comment map according to the second root node and the second leaf node.
In some embodiments, a product review map corresponding to each product is obtained according to each review data. Therefore, the commodity comment map is constructed, commodities evaluated by the user can be displayed better, the emotion scores corresponding to the commodities of the user can be found more visually, and meanwhile, the commodities to be recommended and having higher user evaluation can be determined more conveniently.
Optionally, determining a second feature vector corresponding to the comment data according to the commodity comment map includes: and initializing a second leaf node of the commodity comment map according to the BERT pre-training model to obtain a second feature vector.
Optionally according to presetsThe second algorithm calculates by using the third feature vector to obtain a comment embedded vector corresponding to the commodity, and comprises the following steps: by calculation of
Figure BDA0003347505000000071
Obtaining a comment embedded vector corresponding to a commodity; wherein h isv2Embedding a vector for the comment of the second root node v2, namely embedding a comment vector corresponding to the commodity; n is a radical ofv2Is in the neighborhood of the second root node v 2; h isu2A third eigenvector corresponding to a second leaf node u2 in the neighborhood of the second root node v 2.
Optionally, obtaining an emotion type corresponding to the attribute of the product according to the comment data includes: and inputting the comment data into a preset commodity emotion model to obtain the emotion type corresponding to the attribute of the commodity.
Optionally, the commodity emotion model is obtained according to sample comment data corresponding to the sample commodity and the emotion type corresponding to the attribute of the sample commodity by the user; determining the sample comment data as a training sample, and determining the emotion type corresponding to the attribute of the sample commodity by the user as a training sample label; and training a preset pre-training language model according to the training samples and the training sample labels to obtain a commodity emotion model.
Optionally, the commodity emotion model comprises an extraction model and a classification model; inputting the comment data into a preset commodity emotion model to obtain the emotion types corresponding to the attributes of the commodity, wherein the emotion types comprise: extracting the attribute words in the sample comment data through the extraction model, splicing the sample comment data and the attribute words to obtain input characteristics, and inputting the input characteristics into the classification model to obtain the emotion types corresponding to the attributes of the commodities.
In some embodiments, the commodity emotion model is obtained according to sample comment data corresponding to the sample commodity and emotion types corresponding to attributes of the sample commodity by the user; marking sample comment data by a BIOES (B-begin, I-inside, O-outside, E-end, S-single, named entity marking) method, inputting the marked sample comment data into a BERT pre-training model, obtaining a hidden vector of each word in the marked sample comment data, and extracting by an extraction layer based on a CRF (Conditional Random Field) to obtain attribute words in the sample comment data; splicing the sample comment data and the attribute words to obtain input features, such as [ CLS ] sample comment data [ SEP ] attribute words [ SEP ] ", wherein [ CLS ] is a beginning mark of the input features, and [ SEP ] is a separation mark and an end mark of the input features; and inputting the input features into the BERT layer to obtain feature vectors corresponding to the sample comment data, and inputting the feature vectors corresponding to the sample comment data into the full-link layer to obtain the emotion types corresponding to the attributes of the commodities.
Optionally, determining the to-be-recommended goods according to the goods embedding vectors includes: acquiring the similarity between the embedded vectors of the commodities; and selecting commodities corresponding to the preset number of commodity embedded vectors according to the sequence of the similarity from large to small to determine the commodities to be recommended.
In some embodiments, the articles include article 1, article 2, article 3..... article N, and the article embedding vector for each article includes article embedding vector 1, article embedding vector 2, article embedding vector 3.... article embedding vector N; respectively obtaining the similarity between a commodity embedding vector 1 and a commodity embedding vector 2 and the similarity between the commodity embedding vector 1 and a commodity embedding vector 3. And selecting commodities corresponding to the preset number of commodity embedded vectors according to the sequence of the similarity from large to small to determine the commodities to be recommended.
With reference to fig. 3, an embodiment of the present disclosure provides a method for determining a to-be-recommended product, including:
step S301, obtaining comment data of a plurality of seed users to different commodities and attribute data corresponding to each commodity.
Step S302, obtaining attribute embedded vectors corresponding to the commodities according to the attribute data, and obtaining comment embedded vectors corresponding to the commodities according to the comment data.
And step S303, fusing and splicing the attribute embedded vectors and the comment embedded vectors to obtain commodity embedded vectors corresponding to commodities.
Step S304, the similarity between the embedded vectors of the commodities is obtained.
Step S305, selecting commodities corresponding to the preset number of commodity embedded vectors according to the sequence of the similarity from large to small, and determining the commodities as the commodities to be recommended.
By adopting the method for determining the commodity to be recommended, the attribute embedded vector corresponding to the commodity is obtained through the collected attribute data, and the comment embedded vector corresponding to the commodity is obtained according to the comment data; fusing and splicing the attribute embedded vector and the comment embedded vector corresponding to each commodity to obtain a commodity embedded vector corresponding to each commodity; the similarity between the embedded vectors of the commodities is respectively obtained, and the commodities with the preset number of similarity ranks are determined as the commodities to be recommended, so that the commodities to be recommended which are more suitable for the user preference can be obtained. Meanwhile, by paying attention to the comment of the user on the commodity, the commodity attribute concerned by the user is determined, and the commodity which is similar to the concerned commodity attribute and has higher evaluation is recommended to the user aiming at the concerned point of the user.
Optionally, clustering the users according to the behavior data to obtain a user cluster, including: acquiring a behavior embedding vector corresponding to the behavior data; and calculating by using the behavior embedding vector according to a preset clustering algorithm to obtain the user cluster.
Optionally, the preset clustering algorithm is a K-means clustering algorithm.
Optionally, obtaining a behavior embedding vector corresponding to the behavior data includes: constructing a user behavior map according to each behavior data; and determining a behavior embedding vector corresponding to the behavior data according to the user behavior map.
Optionally, constructing a user behavior map according to each behavior data includes: determining the seed user as a third node, and determining the behavior data corresponding to the seed user as a third leaf node corresponding to the third node; and constructing a user behavior map according to the third node and the third leaf node.
In some embodiments, user behavior maps corresponding to various sub-users are respectively obtained according to various behavior data. Therefore, the user behavior graph is constructed, the user behavior corresponding to the commodity can be better displayed, the operation of the user on the commodity can be more visually found, and meanwhile, the user can be more conveniently clustered, so that a user cluster is obtained.
Optionally, determining a behavior embedding vector corresponding to the behavior data according to the user behavior map includes: initializing a third leaf node of the user behavior map according to a preset mode, obtaining a fourth feature vector corresponding to the third leaf node, and calculating by using the fourth feature vector corresponding to the third leaf node according to a preset third algorithm to respectively obtain behavior embedding vectors corresponding to various sub-users. Alternatively, the preset manner is One-Hot (a form that converts the class variables into a machine learning algorithm that is easy to utilize).
In some embodiments, a fourth feature vector corresponding to the third leaf node is obtained by initializing the third leaf node of the user behavior graph in the form of One-Hot-only coding, for example: s1, S2, S3.. SN, etc.; here, S1 represents purchase behavior, S2 represents collection behavior, S3 represents browsing behavior, and the like.
Optionally, the calculating by using a fourth feature vector corresponding to the third leaf node according to a preset third algorithm to obtain a behavior embedding vector includes: by calculation of
Figure BDA0003347505000000101
Obtaining a behavior embedding vector; wherein S is a behavior embedding vector; n is the total number of commodities; siAnd the fourth feature vector corresponds to the ith third leaf node of the third node.
As shown in fig. 4, an embodiment of the present disclosure provides a method for determining a user cluster, including:
step S401, acquiring behavior data corresponding to a plurality of seed users.
Step S402, acquiring a behavior embedding vector corresponding to the behavior data.
And S403, clustering users by using the behavior embedding vectors according to a preset clustering algorithm to obtain a user cluster.
By adopting the method for determining the user cluster provided by the embodiment of the disclosure, the behavior embedding vector corresponding to the behavior data is obtained by obtaining the behavior data corresponding to the seed user, and the user clustering is performed by using the behavior embedding vector according to the clustering algorithm, so that the users having the same preference can be clustered to obtain the user cluster.
As shown in fig. 5, an embodiment of the present disclosure provides an apparatus for recommending a product, including: an acquisition module 501, a determination module 502 and a recommendation module 503; the obtaining module 501 is configured to obtain comment data of a plurality of seed users on different commodities, attribute data corresponding to each commodity, and behavior data corresponding to each seed user; the determining module 502 is configured to determine a commodity to be recommended according to the attribute data and the comment data, perform user clustering according to the behavior data, and obtain a user cluster; the recommending module 503 is configured to recommend the to-be-recommended product to other users outside the seed user corresponding to the to-be-recommended product in the user cluster.
By adopting the device for recommending commodities provided by the embodiment of the disclosure, the comment data of a plurality of seed users to different commodities, the attribute data corresponding to the commodities and the behavior data corresponding to the seed users are obtained through the obtaining module; the determining module determines commodities to be recommended according to the attribute data and the comment data, and performs user clustering according to the behavior data to obtain a user cluster; and the recommending module recommends the commodity to be recommended to other users outside the seed user corresponding to the commodity to be recommended in the user cluster. Therefore, the commodities to be recommended are determined through the collected comment data and attribute data, the user cluster is obtained through user clustering according to the collected behavior data, the commodities to be recommended are recommended to the users in the user cluster, and commodity recommendation of the crowd related to the users can be achieved.
Optionally, the determining module is configured to determine the goods to be recommended according to the attribute data and the comment data, obtain attribute embedded vectors corresponding to the goods according to the attribute data, and obtain comment embedded vectors corresponding to the goods according to the comment data; fusing and splicing the attribute embedded vectors and the comment embedded vectors to obtain commodity embedded vectors corresponding to commodities; and determining the commodities to be recommended according to the commodity embedded vectors.
Optionally, obtaining an attribute embedded vector corresponding to the commodity according to the attribute data includes: acquiring a first feature vector corresponding to each attribute data; and calculating by using the first feature vector according to a preset first algorithm to obtain an attribute embedded vector corresponding to the commodity.
Optionally, obtaining a comment embedding vector corresponding to the commodity according to the comment data includes: acquiring a second feature vector corresponding to the comment data, and acquiring an emotion score corresponding to the commodity according to the comment data; fusing and splicing the second feature vector and the emotion score to obtain a third feature vector; and calculating by using the third feature vector according to a preset second algorithm to obtain a comment embedded vector corresponding to the commodity.
Optionally, obtaining an emotion score corresponding to the commodity according to the comment data includes: obtaining the emotion type corresponding to the attribute of the commodity according to the comment data; the emotional type includes positive, negative or neutral; counting the emotion types corresponding to the attributes of the commodities to obtain the probability that the emotion types are positive; and determining the probability as the emotion score corresponding to the commodity.
Optionally, determining the to-be-recommended goods according to the goods embedding vectors includes: acquiring the similarity between the embedded vectors of the commodities; and selecting commodities corresponding to the preset number of commodity embedded vectors according to the sequence of the similarity from large to small to determine the commodities to be recommended.
Optionally, the determining module is further configured to perform user clustering according to the behavior data in the following manner, obtain a user cluster, and obtain a behavior embedding vector corresponding to the behavior data; and calculating by using the behavior embedding vector according to a preset clustering algorithm to obtain the user cluster.
In this way, the commodities to be recommended are determined through the collected comment data and attribute data, the user cluster is obtained through user clustering according to the collected behavior data, the commodities to be recommended are recommended to the users in the user cluster, personalized recommendation of the commodities to the crowd related to the users can be achieved according to the recommendation technology, and the commodities recommended to the user cluster meet the preference of all people in the cluster.
As shown in fig. 6, an embodiment of the present disclosure provides an electronic device including a processor (processor)600 and a memory (memory) 601. Optionally, the electronic device may further include a Communication Interface 602 and a bus 603. The processor 600, the communication interface 602, and the memory 601 may communicate with each other via a bus 603. The communication interface 602 may be used for information transfer. The processor 600 may call logic instructions in the memory 601 to perform the method for merchandise recommendation of the above-described embodiment.
By adopting the electronic equipment provided by the embodiment of the disclosure, the comment data of a plurality of seed users on different commodities, the attribute data corresponding to the commodities and the behavior data corresponding to the seed users are obtained; determining commodities to be recommended according to the attribute data and the comment data, and carrying out user clustering according to the behavior data to obtain a user cluster; and recommending the commodity to be recommended to other users outside the seed user corresponding to the commodity to be recommended in the user cluster. Therefore, the commodities to be recommended are determined through the collected comment data and attribute data, the user cluster is obtained through user clustering according to the collected behavior data, the commodities to be recommended are recommended to the users in the user cluster, and commodity recommendation of the crowd related to the users can be achieved.
Optionally, the electronic device includes a server, a computer, a tablet computer, and the like.
In addition, the logic instructions in the memory 601 may be implemented in the form of software functional units and stored in a computer readable storage medium when the logic instructions are sold or used as independent products.
The memory 601 is a computer-readable storage medium, and can be used for storing software programs, computer-executable programs, such as program instructions/modules corresponding to the methods in the embodiments of the present disclosure. The processor 600 executes the functional application and data processing by executing the program instructions/modules stored in the memory 601, that is, implements the method for merchandise recommendation in the above-described embodiment.
The memory 601 may include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function; the storage data area may store data created according to the use of the terminal device, and the like. In addition, the memory 601 may include a high speed random access memory, and may also include a non-volatile memory.
The embodiment of the disclosure provides a storage medium, which stores program instructions, and when the program instructions are executed, the method for recommending commodities is executed.
The disclosed embodiments provide a computer program product comprising a computer program stored on a computer readable storage medium, the computer program comprising program instructions which, when executed by a computer, cause the computer to perform the above-described method for merchandise recommendation.
The computer-readable storage medium described above may be a transitory computer-readable storage medium or a non-transitory computer-readable storage medium.
The technical solution of the embodiments of the present disclosure may be embodied in the form of a software product, where the computer software product is stored in a storage medium and includes one or more instructions to enable a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method of the embodiments of the present disclosure. And the aforementioned storage medium may be a non-transitory storage medium comprising: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes, and may also be a transient storage medium.
The above description and drawings sufficiently illustrate embodiments of the disclosure to enable those skilled in the art to practice them. Other embodiments may incorporate structural, logical, electrical, process, and other changes. The examples merely typify possible variations. Individual components and functions are optional unless explicitly required, and the sequence of operations may vary. Portions and features of some embodiments may be included in or substituted for those of others. Furthermore, the words used in the specification are words of description only and are not intended to limit the claims. As used in the description of the embodiments and the claims, the singular forms "a", "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. Similarly, the term "and/or" as used in this application is meant to encompass any and all possible combinations of one or more of the associated listed. Furthermore, the terms "comprises" and/or "comprising," when used in this application, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof. Without further limitation, an element defined by the phrase "comprising an …" does not exclude the presence of other like elements in a process, method or apparatus that comprises the element. In this document, each embodiment may be described with emphasis on differences from other embodiments, and the same and similar parts between the respective embodiments may be referred to each other. For methods, products, etc. of the embodiment disclosures, reference may be made to the description of the method section for relevance if it corresponds to the method section of the embodiment disclosure.
Those of skill in the art would appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software may depend upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the disclosed embodiments. It can be clearly understood by the skilled person that, for convenience and brevity of description, the specific working processes of the system, the apparatus and the unit described above may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
In the embodiments disclosed herein, the disclosed methods, products (including but not limited to devices, apparatuses, etc.) may be implemented in other ways. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the units may be merely a logical division, and in actual implementation, there may be another division, for example, multiple units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form. The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to implement the present embodiment. In addition, functional units in the embodiments of the present disclosure may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit.
The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). In some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. In the description corresponding to the flowcharts and block diagrams in the figures, operations or steps corresponding to different blocks may also occur in different orders than disclosed in the description, and sometimes there is no specific order between the different operations or steps. For example, two sequential operations or steps may in fact be executed substantially concurrently, or they may sometimes be executed in the reverse order, depending upon the functionality involved. Each block of the block diagrams and/or flowchart illustrations, and combinations of blocks in the block diagrams and/or flowchart illustrations, can be implemented by special purpose hardware-based systems that perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.

Claims (10)

1. A method for merchandise recommendation, comprising:
the method comprises the steps of obtaining comment data of a plurality of seed users on different commodities, attribute data corresponding to the commodities and behavior data corresponding to the seed users;
determining a commodity to be recommended according to the attribute data and the comment data, and performing user clustering according to the behavior data to obtain a user cluster;
recommending the commodity to be recommended to other users outside the seed user corresponding to the commodity to be recommended in the user cluster.
2. The method of claim 1, wherein determining the item to be recommended from the attribute data and the comment data comprises:
acquiring attribute embedded vectors corresponding to the commodities according to the attribute data, and acquiring comment embedded vectors corresponding to the commodities according to the comment data;
fusing and splicing each attribute embedded vector and each comment embedded vector to obtain a commodity embedded vector corresponding to each commodity;
and determining the commodities to be recommended according to the commodity embedded vectors.
3. The method of claim 2, wherein obtaining the attribute embedded vector corresponding to the commodity according to the attribute data comprises:
acquiring a first feature vector corresponding to the attribute data;
and calculating by using the first feature vector according to a preset first algorithm to obtain an attribute embedded vector corresponding to the commodity.
4. The method of claim 2, wherein obtaining a comment embedded vector corresponding to the commodity according to the comment data comprises:
acquiring a second feature vector corresponding to the comment data, and acquiring an emotion score corresponding to the commodity according to the comment data;
fusing and splicing the second feature vector and the emotion score to obtain a third feature vector;
and calculating by using the third feature vector according to a preset second algorithm to obtain a comment embedded vector corresponding to the commodity.
5. The method of claim 4, wherein obtaining the sentiment score corresponding to the commodity according to the comment data comprises:
obtaining the emotion type corresponding to the attribute of the commodity according to the comment data; the emotion types include positive, negative or neutral;
counting the emotion types corresponding to the attributes of the commodities to obtain the probability that the emotion types are positive;
and determining the probability as the emotion score corresponding to the commodity.
6. The method of claim 2, wherein determining the items to be recommended based on each of the item embedding vectors comprises:
acquiring the similarity between the embedded vectors of the commodities;
and selecting a preset number of commodities corresponding to the commodity embedding vectors according to the sequence of the similarity from large to small to determine the commodities to be recommended.
7. The method according to any one of claims 1 to 6, wherein clustering users according to the behavior data to obtain user clusters comprises:
acquiring a behavior embedding vector corresponding to the behavior data;
and calculating by utilizing the behavior embedding vector according to a preset clustering algorithm to obtain the user cluster.
8. An apparatus for merchandise recommendation, comprising:
the acquisition module is configured to acquire comment data of a plurality of seed users on different commodities, attribute data corresponding to the commodities and behavior data corresponding to the seed users;
the determining module is configured to determine commodities to be recommended according to the attribute data and the comment data, and perform user clustering according to the behavior data to obtain a user cluster;
and the recommending module is configured to recommend the commodity to be recommended to other users outside the seed user corresponding to the commodity to be recommended in the user cluster.
9. An electronic device comprising a processor and a memory storing program instructions, wherein the processor is configured to perform the method of any of claims 1 to 7 when executing the program instructions.
10. A storage medium storing program instructions, characterized in that said program instructions, when executed, perform a method for merchandise recommendation according to any one of claims 1 to 7.
CN202111327653.7A 2021-11-10 2021-11-10 Method and device for commodity recommendation, electronic equipment and storage medium Pending CN114090880A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111327653.7A CN114090880A (en) 2021-11-10 2021-11-10 Method and device for commodity recommendation, electronic equipment and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111327653.7A CN114090880A (en) 2021-11-10 2021-11-10 Method and device for commodity recommendation, electronic equipment and storage medium

Publications (1)

Publication Number Publication Date
CN114090880A true CN114090880A (en) 2022-02-25

Family

ID=80299609

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111327653.7A Pending CN114090880A (en) 2021-11-10 2021-11-10 Method and device for commodity recommendation, electronic equipment and storage medium

Country Status (1)

Country Link
CN (1) CN114090880A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115098650A (en) * 2022-08-25 2022-09-23 华扬联众数字技术股份有限公司 Comment information analysis method based on historical data model and related device
CN116628345A (en) * 2023-07-13 2023-08-22 腾讯科技（深圳）有限公司 Content recommendation method and device, electronic equipment and storage medium

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115098650A (en) * 2022-08-25 2022-09-23 华扬联众数字技术股份有限公司 Comment information analysis method based on historical data model and related device
CN116628345A (en) * 2023-07-13 2023-08-22 腾讯科技（深圳）有限公司 Content recommendation method and device, electronic equipment and storage medium
CN116628345B (en) * 2023-07-13 2024-02-06 腾讯科技（深圳）有限公司 Content recommendation method and device, electronic equipment and storage medium

Similar Documents

Publication Publication Date Title
CN108959603B (en) Personalized recommendation system and method based on deep neural network
CN111784455B (en) Article recommendation method and recommendation equipment
CN106599226B (en) Content recommendation method and content recommendation system
CN107833082B (en) Commodity picture recommendation method and device
CN108829822A (en) The recommended method and device of media content, storage medium, electronic device
CN114090880A (en) Method and device for commodity recommendation, electronic equipment and storage medium
US11599927B1 (en) Artificial intelligence system using deep neural networks for pairwise character-level text analysis and recommendations
CN107862322B (en) Method, device and system for classifying picture attributes by combining picture and text
CN110580489B (en) Data object classification system, method and equipment
CN110674620A (en) Target file generation method, device, medium and electronic equipment
CN112948575B (en) Text data processing method, apparatus and computer readable storage medium
CN111966888B (en) Aspect class-based interpretability recommendation method and system for fusing external data
CN111667337A (en) Commodity evaluation ordering method and system
CN114049174A (en) Method and device for commodity recommendation, electronic equipment and storage medium
CN115577095B (en) Electric power standard information recommendation method based on graph theory
CN111523315B (en) Data processing method, text recognition device and computer equipment
CN116703506A (en) Multi-feature fusion-based E-commerce commodity recommendation method and system
CN110738050A (en) Text recombination method, device and medium based on word segmentation and named entity recognition
CN115641179A (en) Information pushing method and device and electronic equipment
CN110851694A (en) Personalized recommendation system based on user memory network and tree structure depth model
CN112800226A (en) Method for obtaining text classification model, method, device and equipment for text classification
CN109948040A (en) Storage, recommended method and the system of object information, equipment and storage medium
CN111597296A (en) Commodity data processing method, device and system
Gaurav et al. Detection of false positive situation in review mining
CN108763258A (en) Document subject matter parameter extracting method, Products Show method, equipment and 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