CN112330382A - Item recommendation method and device, computing equipment and medium - Google Patents
Item recommendation method and device, computing equipment and medium Download PDFInfo
- Publication number
- CN112330382A CN112330382A CN202010471229.9A CN202010471229A CN112330382A CN 112330382 A CN112330382 A CN 112330382A CN 202010471229 A CN202010471229 A CN 202010471229A CN 112330382 A CN112330382 A CN 112330382A
- Authority
- CN
- China
- Prior art keywords
- target
- historical
- model
- search
- item
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 51
- 238000012545 processing Methods 0.000 claims abstract description 39
- 238000012549 training Methods 0.000 claims abstract description 34
- 230000015654 memory Effects 0.000 claims description 10
- 230000004044 response Effects 0.000 claims description 4
- 238000010586 diagram Methods 0.000 description 12
- 238000004422 calculation algorithm Methods 0.000 description 11
- 238000004590 computer program Methods 0.000 description 11
- 230000008569 process Effects 0.000 description 8
- 238000004891 communication Methods 0.000 description 7
- 230000000694 effects Effects 0.000 description 7
- 230000006870 function Effects 0.000 description 7
- 235000013399 edible fruits Nutrition 0.000 description 6
- 238000004364 calculation method Methods 0.000 description 5
- 239000010749 BS 2869 Class C1 Substances 0.000 description 4
- 239000010750 BS 2869 Class C2 Substances 0.000 description 4
- 238000002372 labelling Methods 0.000 description 3
- 241000234671 Ananas Species 0.000 description 2
- 235000007119 Ananas comosus Nutrition 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000004806 packaging method and process Methods 0.000 description 2
- 239000000758 substrate Substances 0.000 description 2
- 241001164374 Calyx Species 0.000 description 1
- 108010001267 Protein Subunits Proteins 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 239000000835 fiber Substances 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0631—Item recommendations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/35—Clustering; Classification
-
- 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
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Finance (AREA)
- Accounting & Taxation (AREA)
- Strategic Management (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Development Economics (AREA)
- Databases & Information Systems (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Economics (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Game Theory and Decision Science (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The present disclosure provides an item recommendation method, including: acquiring a target search word of a user; processing the target search terms by using the target model to obtain target articles associated with the target search terms, and recommending the target articles to the user; wherein the target model is obtained based on: obtaining a plurality of first historical search terms, wherein each of the plurality of first historical search terms has label information, and the label information represents historical articles associated with the first historical search terms; and training the initial model by using the plurality of first historical search words to obtain a target model, wherein the initial model is obtained by clustering the plurality of second historical search words. The disclosure also provides an item recommendation apparatus, a computing device and a computer-readable storage medium.
Description
Technical Field
The present disclosure relates to the field of computer technologies, and in particular, to an item recommendation method, an item recommendation apparatus, a computing device, and a computer-readable storage medium.
Background
Related art typically matches related items for a user based on the user's search terms. For example, the e-commerce platform may recommend related goods (items) to the user based on the user's search terms. When a user needs a certain commodity, the user can input a search word in a search interface provided by the e-commerce platform, and the e-commerce platform performs matching calculation according to the search word of the user to obtain a related commodity so as to recommend the related commodity to the user. The matching calculation is, for example, to match a search term input by a user with a name of a commodity.
In carrying out the presently disclosed concept, the inventors have found that there are at least the following problems in the related art.
In the related art, when related goods are obtained by performing matching calculation according to a search term input by a user, when the search term input by the user is rare or wrong, the goods required by the user cannot be obtained by performing matching calculation, so that the use experience of the user is poor.
Disclosure of Invention
In view of the above, the present disclosure provides an optimized item recommendation method, an item recommendation apparatus, a computing device, and a computer-readable storage medium.
One aspect of the present disclosure provides an item recommendation method, including: the method comprises the steps of obtaining a target search word of a user, processing the target search word by using a target model to obtain a target article associated with the target search word, and recommending the target article to the user. Wherein the target model is obtained based on: the method comprises the steps of obtaining a plurality of first historical search terms, wherein each first historical search term in the plurality of first historical search terms respectively has label information, the label information represents historical articles related to the first historical search terms, and an initial model is trained by the plurality of first historical search terms to obtain a target model, wherein the initial model is obtained by clustering a plurality of second historical search terms.
According to an embodiment of the present disclosure, the training of the initial model by using the plurality of first history search terms includes: and when the initial model is used for clustering each first historical search word in the plurality of first historical search words, adjusting model parameters of the initial model based on label information of each first historical search word so that a first clustering result obtained by clustering corresponds to the label information.
According to an embodiment of the present disclosure, the first clustering result includes a plurality of first categories, and each of the plurality of first categories characterizes at least one first item. Wherein the processing the target search term by using the target model to obtain the target item associated with the target search term comprises: clustering the target search terms by using the target model to determine a target category to which the target search terms belong in the plurality of first categories; and
determining that the at least one first item characterized by the target category is the target item.
According to an embodiment of the present disclosure, the method further includes: obtaining a model to be trained, and performing clustering processing on the plurality of second historical search terms by using the model to be trained to obtain the initial model, wherein a second clustering result obtained through the clustering processing comprises a plurality of second categories, each of the plurality of second categories comprises at least one second historical search term, and for each of the plurality of second categories, second articles associated with each of the at least one second historical search term are the same.
According to an embodiment of the present disclosure, the obtaining a plurality of first history search terms includes: determining the order placing time of a user for purchasing each historical item in a plurality of historical items, acquiring a plurality of third historical search terms in a preset time period before the order placing time based on the order placing time for each historical item, and determining at least one third historical search term associated with the historical item in the plurality of third historical search terms as the first historical search term.
According to an embodiment of the present disclosure, the method further includes: and associating each first historical search word in the plurality of first historical search words with the historical item corresponding to each first historical search word so that each first historical search word has the label information.
According to an embodiment of the present disclosure, the processing the target search term by using the target model includes: and determining whether the associated articles related to the target search word can be obtained by matching in a matching mode, and processing the target search word by using the target model in response to determining that the associated articles related to the target search word cannot be matched in the matching mode.
Another aspect of the present disclosure provides an item recommendation device including: the device comprises a first acquisition module, a processing module, a second acquisition module and a training module. The first obtaining module obtains a target search word of a user. And the processing module is used for processing the target search terms by utilizing a target model to obtain target articles associated with the target search terms and recommending the target articles to the user. Wherein the target model is obtained based on: the second obtaining module obtains a plurality of first historical search terms, wherein each of the plurality of first historical search terms has label information, and the label information represents historical items associated with the first historical search terms. And the training module is used for training an initial model by using the plurality of first historical search words to obtain the target model, wherein the initial model is obtained by clustering a plurality of second historical search words.
According to an embodiment of the present disclosure, the training of the initial model by using the plurality of first history search terms includes: and when the initial model is used for clustering each first historical search word in the plurality of first historical search words, adjusting model parameters of the initial model based on label information of each first historical search word so that a first clustering result obtained by clustering corresponds to the label information.
According to an embodiment of the present disclosure, the first clustering result includes a plurality of first categories, and each of the plurality of first categories characterizes at least one first item. Wherein the processing the target search term by using the target model to obtain the target item associated with the target search term comprises: clustering the target search terms by using the target model to determine a target category to which the target search terms belong in the plurality of first categories; and
determining that the at least one first item characterized by the target category is the target item.
According to the embodiment of the present disclosure, the apparatus further includes: a third obtaining module and a clustering module. And the third acquisition module acquires the model to be trained. And the clustering module is used for clustering the plurality of second historical search terms by using the model to be trained to obtain the initial model, wherein a second clustering result obtained by clustering comprises a plurality of second categories, each second category in the plurality of second categories comprises at least one second historical search term, and for each category in the plurality of second categories, second articles associated with each second historical search term in the at least one second historical search term are the same.
According to an embodiment of the present disclosure, the obtaining a plurality of first history search terms includes: determining the order placing time of a user for purchasing each historical item in a plurality of historical items, acquiring a plurality of third historical search terms in a preset time period before the order placing time based on the order placing time for each historical item, and determining at least one third historical search term associated with the historical item in the plurality of third historical search terms as the first historical search term.
According to the embodiment of the present disclosure, the apparatus further includes: and the association module is used for associating each first historical search word in the plurality of first historical search words with the historical item corresponding to each first historical search word so as to enable each first historical search word to have the label information.
According to an embodiment of the present disclosure, the processing the target search term by using the target model includes: and determining whether the associated articles related to the target search word can be obtained by matching in a matching mode, and processing the target search word by using the target model in response to determining that the associated articles related to the target search word cannot be matched in the matching mode.
Another aspect of the present disclosure provides a computer-readable storage medium storing computer-executable instructions for implementing the method as described above when executed.
Another aspect of the disclosure provides a computer program comprising computer executable instructions for implementing the method as described above when executed.
According to the embodiment of the disclosure, the problem that in the related art, when the search word input by the user is rare or wrong, the user experience is not good due to the fact that the article required by the user cannot be obtained through matching calculation can be solved at least partially by using the article recommending method or device, and therefore the related articles can be recommended for the user according to the association between the historical search word and the articles in the historical order, the article recommending accuracy is improved, and the technical effect of the user experience is improved.
Drawings
The above and other objects, features and advantages of the present disclosure will become more apparent from the following description of embodiments of the present disclosure with reference to the accompanying drawings, in which:
fig. 1 schematically illustrates a system architecture of an item recommendation method and apparatus according to an embodiment of the present disclosure;
2A-2B schematically illustrate a flow diagram of an item recommendation method performed by a computing device, according to an embodiment of the present disclosure;
FIG. 2C schematically illustrates an initial model training diagram according to an embodiment of the disclosure;
3A-3B schematically illustrate block diagrams of an item recommendation device according to an embodiment of the present disclosure; and
FIG. 4 schematically illustrates a block diagram of a computer system adapted for item recommendation, in accordance with an embodiment of the present disclosure.
Detailed Description
Hereinafter, embodiments of the present disclosure will be described with reference to the accompanying drawings. It should be understood that the description is illustrative only and is not intended to limit the scope of the present disclosure. In the following detailed description, for purposes of explanation, numerous specific details are set forth in order to provide a thorough understanding of the embodiments of the disclosure. It may be evident, however, that one or more embodiments may be practiced without these specific details. Moreover, in the following description, descriptions of well-known structures and techniques are omitted so as to not unnecessarily obscure the concepts of the present disclosure.
The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the disclosure. The terms "comprises," "comprising," and the like, as used herein, specify the presence of stated features, steps, operations, and/or components, but do not preclude the presence or addition of one or more other features, steps, operations, or components.
All terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art unless otherwise defined. It is noted that the terms used herein should be interpreted as having a meaning that is consistent with the context of this specification and should not be interpreted in an idealized or overly formal sense.
Where a convention analogous to "at least one of A, B and C, etc." is used, in general such a construction is intended in the sense one having skill in the art would understand the convention (e.g., "a system having at least one of A, B and C" would include but not be limited to systems that have a alone, B alone, C alone, a and B together, a and C together, B and C together, and/or A, B, C together, etc.).
An embodiment of the present disclosure provides an item recommendation method performed by a computing device, including: and acquiring a target search word of the user, processing the target search word by using a target model to obtain a target article associated with the target search word, and recommending the target article to the user. Wherein the target model is obtained based on: the method comprises the steps of obtaining a plurality of first historical search terms, wherein each first historical search term in the plurality of first historical search terms respectively has label information, and the label information represents historical articles related to the first historical search terms. And then, training an initial model by using the plurality of first historical search words to obtain a target model, wherein the initial model is obtained by clustering a plurality of second historical search words.
Fig. 1 schematically shows a system architecture of an item recommendation method and apparatus according to an embodiment of the present disclosure. It should be noted that fig. 1 is only an example of a system architecture to which the embodiments of the present disclosure may be applied to help those skilled in the art understand the technical content of the present disclosure, and does not mean that the embodiments of the present disclosure may not be applied to other devices, systems, environments or scenarios.
As shown in fig. 1, the system architecture 100 according to this embodiment may include terminal devices 101, 102, 103, a network 104 and a server 105. The network 104 serves as a medium for providing communication links between the terminal devices 101, 102, 103 and the server 105. Network 104 may include various connection types, such as wired, wireless communication links, or fiber optic cables, to name a few.
The user may use the terminal devices 101, 102, 103 to interact with the server 105 via the network 104 to receive or send messages or the like. The terminal devices 101, 102, 103 may have installed thereon various communication client applications, such as shopping-like applications, web browser applications, search-like applications, instant messaging tools, mailbox clients, social platform software, etc. (by way of example only).
The terminal devices 101, 102, 103 may be various electronic devices having a display screen and supporting web browsing, including but not limited to smart phones, tablet computers, laptop portable computers, desktop computers, and the like.
The server 105 may be a server providing various services, such as a background management server (for example only) providing support for websites browsed by users using the terminal devices 101, 102, 103. The background management server may analyze and perform other processing on the received data such as the user request, and feed back a processing result (e.g., a webpage, information, or data obtained or generated according to the user request) to the terminal device.
It should be noted that the item recommendation method provided by the embodiment of the present disclosure may be generally executed by the server 105. Accordingly, the item recommendation device provided by the embodiments of the present disclosure may be generally disposed in the server 105. The item recommendation method provided by the embodiment of the present disclosure may also be executed by a server or a server cluster different from the server 105 and capable of communicating with the terminal devices 101, 102, 103 and/or the server 105. Accordingly, the item recommendation device provided by the embodiment of the present disclosure may also be disposed in a server or a server cluster different from the server 105 and capable of communicating with the terminal devices 101, 102, 103 and/or the server 105.
For example, in the embodiment of the present disclosure, a target search term input by a user may be received by the terminal devices 101, 102, 103 and stored in the terminal devices 101, 102, 103, and the target search term is sent to the server 105 through the terminal devices 101, 102, 103, and the server 105 may process the target search term by using a target model to obtain a target item associated with the target search term. Fig. 1 shows an example in which the terminal apparatus 101 receives a target search word, for example.
For example, the first and second historical search terms of the embodiment of the present disclosure may be directly stored in the server 105, the initial model is trained by the server 105 directly using the plurality of first historical search terms to obtain the target model, and the initial model is obtained by the server 105 directly performing clustering based on the plurality of second historical search terms.
It should be understood that the number of terminal devices, networks, and servers in fig. 1 is merely illustrative. There may be any number of terminal devices, networks, and servers, as desired for implementation.
Fig. 2A-2B schematically illustrate a flow diagram of an item recommendation method performed by a computing device, according to an embodiment of the disclosure.
As shown in fig. 2A, the item recommendation method of embodiments of the present disclosure may be performed by a computing device, for example. Computing devices may include, but are not limited to, smart phones, tablets, laptop portable computers, desktop computers, and servers. The method may include, for example, the following operations S210 to S220.
In operation S210, a target search term of a user is acquired.
According to the embodiment of the disclosure, the target search word may be a search word input by a user on a search interface of the e-commerce platform, for example. The user searches for a desired item by inputting a target search term, and the item of the embodiment of the present disclosure may be, for example, a commodity provided by an e-commerce platform.
Next, in operation S220, the target search term is processed using the target model, a target item associated with the target search term is obtained, and the target item is recommended to the user.
In the disclosed embodiment, the computing device of the disclosed embodiment may be, for example, a computing device in an e-commerce platform, or the computing device of the disclosed embodiment may be the server 105 shown in fig. 1. The related functions described in the embodiments of the present disclosure as being performed by the e-commerce platform may each be performed by a computing device, for example.
According to the embodiment of the disclosure, after receiving the target search terms of the user, the e-commerce platform can recommend the target articles associated with the target search terms for the user according to the target search terms. For example, the target search term may be processed using a target model to obtain the target item. The target model is, for example, a trained clustering model, and a plurality of clustering results can be obtained after clustering is performed by using the target model, and each clustering result in the plurality of clustering results represents, for example, an article. When the target search term is input into the target model, the target model outputs, for example, a probability that the target search term belongs to each of the plurality of clustered results.
For example, with 3 clustering resultsFor example, clustering result A1Clustering result A2Clustering result A3For example to characterize the articles a separately1Article a2Article a3. After the target search words are clustered by using the target model, the target search words are obtained and are respectively associated with the articles a1Article a2Article a3And the associated probability can determine the object with the high probability as the target object. For example, when the target search word is processed by the target model to obtain the probability that the target search word belongs to each cluster result, the target search word respectively belongs to the cluster result A1Clustering result A2Clustering result A3For example, the probability of (a) represents the target search term and the object (a) respectively1Article a2Article a3The associated probability. Wherein, when the target search word is related to the article a1The associated probability is greater than that of the target search word and the article a2And/or article a3When the associated probability is reached, the target item associated with the target search term can be determined to be the item a1And article a1And recommending to the user. The manner of obtaining the target model will be described below with reference to fig. 2B.
In an embodiment, after the target search term input by the user is obtained, it may be first determined whether the associated item related to the target search term can be obtained by matching using a matching method, and if it is determined that the associated item related to the target search term cannot be obtained by matching using the matching method, the target search term is further processed using the target model to obtain the target item.
The matching mode may be, for example, matching the target search term with the item name of the item in the e-commerce platform, and if the similarity between the target search term and the item name of the associated item is greater than a preset threshold, recommending the associated item for the user.
If the associated articles are not matched in a matching mode, the similarity degree between the target search words input by the user and the article names is small, and the target search words input by the user can be wrong search words or rare search words. When the associated item is not matched in a matching manner, the target search term can be processed by using the target model so as to recommend the target item for the user.
Aiming at the condition that the search word input by the user is the wrong search word. For example, when a user wants to search for "basketball", if the input error is "blue quest", the item "basketball" is not usually matched by matching.
Aiming at the condition that the search word input by the user is an uncommon search word. For example, when the item required by the user is "calyx fruit", the target search word "pineapple fruit" input by the user is caused due to geographical awareness. "pineapple fruit" is, for example, the name of the girl fruit in some areas only, but not the general name of the "girl fruit", and the item "girl fruit" which is not usually needed by the user is matched through a matching mode.
As shown in fig. 2B, the target model of the embodiment of the present disclosure may be obtained based on, for example, the following operations S230 to S240.
In operation S230, a plurality of first historical search terms are obtained, wherein each of the plurality of first historical search terms has tag information, and the tag information represents a historical item associated with the first historical search term.
According to an embodiment of the present disclosure, for example, a time to order for a user to purchase each historical item of a plurality of historical items is determined. For each historical item, a plurality of third historical search terms within a preset time period before the order-placing time are acquired based on the order-placing time. The preset time period may be, for example, 3 days, 4 days, 5 days, etc. After the user performs the order placing operation on one historical item, all search terms within 3 days before the order placing time are taken as third historical search terms. For example, the commodity of the order made by the user is "basketball", and the plurality of third history search terms within 3 days before the order made by the user include "blue quest", "mobile phone", "computer", and the like.
Then, at least one third history search word associated with the historical item in the plurality of third history search words is determined as the first history search word. For example, the third history search term "lan quest" may be used as the first history search term.
According to the embodiment of the disclosure, each first historical search word in a plurality of first historical search words and the historical item corresponding to each first historical search word are associated, so that each first historical search word has label information. For example, the first historical search term "blue quest" is associated with the item "basketball," such that the label information of the first historical search term "blue quest" is "basketball.
Next, in operation S240, an initial model is trained using the plurality of first historical search terms to obtain a target model, wherein the initial model is obtained by clustering a plurality of second historical search terms, for example.
According to the embodiment of the disclosure, the initial model is, for example, a clustering model trained by using a plurality of second historical search terms. Each of the plurality of second historical search terms, for example, has no tag information. For example, the initial model includes an LDA (Latent Dirichlet Allocation) algorithm model, and the training process of the initial model is, for example, an unsupervised training process, that is, the training data (second historical search terms) does not need to have label information, so that labeling for each second historical search term is not needed, and the labeling cost is reduced. Wherein the training process of the initial model will be described below.
Then, each first historical search word in the plurality of first historical search words is clustered by using the initial model. When each of the plurality of first historical search terms is clustered by using the initial model, the model parameters of the initial model may be adjusted based on the tag information of each of the first historical search terms, so that the first clustering result obtained by clustering corresponds to the tag information.
According to the embodiment of the disclosure, after the initial model is obtained through training, the initial model can be supervised and trained again based on the first historical search word with the label information, so that the clustering effect of the model is improved. When the initial model is trained by using the first historical search word, the clustering result of the initial model can be adjusted according to the label information of the first historical search word, the model parameter of the initial model is adjusted reversely according to the clustering result, the initial model after the model parameter is adjusted is used for continuously clustering the first historical search word to obtain the clustering result, and the model parameter is adjusted reversely on the basis of the clustering result until the initial model reaches the convergence state. When the initial model reaching the convergence state is used for clustering the first historical search term, the obtained first clustering result is consistent with the label information of the first historical search term, for example. For example, when the first historical search term "blue quest" is clustered using the initial model of the convergence state, the first clustering result, for example, characterizes that the probability that "blue quest" belongs to the item "basketball" is greater than the probability that "blue quest" belongs to other items. The converged initial model serves as a target model for embodiments of the present disclosure, for example.
According to an embodiment of the present disclosure, the target search term may be a search term input by a first user, the second history search term may be a search term input by a second user, and the third history search term may be a search term input by a third user. The first user may include one or more users, the second user may include one or more users, and the third user may also include one or more users.
In an embodiment, the first user, the second user, and the third user may be different users. For example, the first user comprises user A and the second user comprises user B1User B2… …. The third user includes user C1User C2、……。
In another embodiment, the first user, the second user, and the third user may have partially identical users with respect to each other. For example, the first user comprises user A and the second user comprises user B1User B2User A, user C1… … are provided. The third user includes user C1User C2User A, user B2……。
It can be understood that the embodiment of the present disclosure obtains the first historical search term with the tag information by associating the first historical search term with the ordered historical item. Then, the initial model is trained by using the first historical search word with the label information to obtain a target model, so that a target item associated with the target search word is recommended to the user by using the target model according to the target search word input by the user. Therefore, by the technical scheme of the embodiment of the disclosure, the object really needed by the user can still be known even if the target search word input by the user is rare or wrong, so that the object recommendation accuracy is improved, and the use experience of the user is improved.
In addition, the number of the second history search terms of the embodiment of the present disclosure is far greater than the number of the first history search terms. And carrying out unsupervised training on the second historical search terms without label information to obtain an initial model, and reducing the cost of labeling each second historical search term in an unsupervised training mode. Then, supervised training is carried out based on the first historical search word with the label information to obtain a target model, and the clustering effect of the target model is improved. Therefore, the final target model is obtained in a semi-supervised mode, and the training effect and the clustering effect of the model are improved.
According to the embodiment of the present disclosure, the first clustering result may include, for example, a result obtained by clustering the target search term using a target model, where the target model is a converged initial model obtained by training an initial model using the first historical search term. The first clustering result includes, for example, a plurality of first categories, each of which characterizes, for example, at least one first item. For example, the first clustering result includes a first category B1Class B2Class B3First class B1Class B2Class B3For example to characterize the first article b separately1A first article b2A first article b3。
In the operation S220, for example, the obtaining of the target item associated with the target search term by processing the target search term using the target model may include: and clustering the target search terms by using a target model to determine a target category to which the target search terms belong in a plurality of first categories, and then determining at least one first article represented by the target category as a target article.
For example, when the target search word is "blue solution", the target search word "blue solution" is clustered by using the target model to determine a target category to which the target search word "blue solution" belongs among the plurality of first categories. For example, when the object class is the first class B1When in the first class B1Characterized first article b1For example, "basketball," the item "basketball" is the target item associated with the target search term "blue quest.
According to an embodiment of the present disclosure, the initial model acquisition process is described as follows, for example.
Firstly, a model to be trained is obtained, and the model to be trained is utilized to perform clustering processing on a plurality of second historical search terms so as to obtain an initial model.
And for each category in the plurality of second categories, the second items associated with each second history search word in the at least one second history search word are the same.
Fig. 2C schematically illustrates an initial model training diagram according to an embodiment of the disclosure.
As shown in FIG. 2C, the plurality of second historical search terms may include, for example, search term d1Search term d2Search term d3… …, search term dnN is, for example, an integer greater than 1. Clustering the plurality of second history search terms, for example, to obtain m second categories, where the m categories are, for example, integers greater than 1. Each second category includes, for example, at least one second historical search term. For example, in a second class C1The second class C2For example, the second class C1For example, characterise the second object c1Of the second class C2For example, characterise the second object c2. Belong to a second class C1For example, includes search term d1Search term d2Etc., belonging to a second class C2Second historical search terms such as a packageIncluding search term d3Search term d4Search term d5And so on. Wherein it belongs to the second class C1Search term d of1Search term d2All related articles such as the second article c1Belong to a second class C2Search term d of3Search term d4Search term d5All related articles such as the second article c2。
According to the embodiment of the present disclosure, since the initial model is an lda (latent Dirichlet allocation) algorithm model, in the process of training the model to be trained by using the second historical search word to obtain the initial model, since the input of the model to be trained is generally a plurality of document data, each document data includes a plurality of word data. Therefore, the plurality of second history search words may be divided into a plurality of groups, each group being, for example, one document data, and the word data included in each document data being, for example, the second history search words in the group. Then, a plurality of document data are input into the model to be trained for training to obtain an initial model.
According to the embodiment of the disclosure, if the search information input by the user is a sentence, word segmentation processing can be performed on the sentence to obtain a first historical search word, a second historical search word or a target search word. In addition, the search information input by the user may also be voice information, and the first history search word, the second history search word or the target search word may be obtained by performing voice processing on the voice information.
The embodiment of the present disclosure takes, for example, an invalid search word including, for example, a search word input in error or an unusual search word as the first history search word. Meanwhile, the ordering behavior of the user is recorded, the ordering article is used for marking the first history search word input before to obtain the first history search word with the label information, the LDA algorithm model is trained by utilizing the first history search word with the label information, clustering of the first history search word is achieved, and when the subsequent user inputs the similar or same first history search word, the target article related to the first history search word can be actively recommended to the user. Therefore, even under the condition that the user inputs invalid search words, the article really needed by the user can be predicted with a high probability, and the article recommending effect is improved.
The clustering model of the embodiment of the present disclosure is, for example, a semi-supervised LDA algorithm model, which is, for example, an optimization algorithm combining supervised data on the basis of an unsupervised LDA algorithm model. Because the LDA algorithm model is a completely unsupervised clustering algorithm, in order to improve the accuracy of clustering, the LDA algorithm is trained by the first history search word with the label information to obtain the semi-supervised LDA algorithm model. The semi-supervised LDA algorithm model not only has the advantage of good clustering effect, but also reduces the marking cost of training data.
Fig. 3A-3B schematically illustrate block diagrams of an item recommendation device according to an embodiment of the present disclosure.
As shown in fig. 3A, the item recommendation device 300 may include, for example, a first obtaining module 310 and a processing module 320.
The first obtaining module 310 may be used to obtain a target search term of a user. According to the embodiment of the present disclosure, the first obtaining module 310 may, for example, perform operation S210 described above with reference to fig. 2A, which is not described herein again.
The processing module 320 may be configured to process the target search term using the target model, obtain a target item associated with the target search term, and recommend the target item to the user. According to the embodiment of the present disclosure, the processing module 320 may, for example, perform operation S220 described above with reference to fig. 2A, which is not described herein again.
As shown in fig. 3B, the item recommendation device 300 includes, for example, a second obtaining module 330 and a training module 340, in addition to the first obtaining module 310 and the processing module 320. Wherein the target model is obtained based on the second obtaining module 330 and the training module 340.
The second obtaining module 330 may be configured to obtain a plurality of first historical search terms, where each of the plurality of first historical search terms has tag information, and the tag information represents a historical item associated with the first historical search term. According to the embodiment of the present disclosure, the second obtaining module 330 may, for example, perform operation S230 described above with reference to fig. 2B, which is not described herein again.
The training module 340 may be configured to train an initial model with a plurality of first historical search terms to obtain a target model, where the initial model is obtained by clustering a plurality of second historical search terms. According to an embodiment of the present disclosure, the training module 340 may, for example, perform the operation S240 described above with reference to fig. 2B, which is not described herein again.
According to an embodiment of the present disclosure, the training the initial model by using the plurality of first history search words includes: and when the initial model is used for clustering each first historical search word in the plurality of first historical search words, adjusting model parameters of the initial model based on the label information of each first historical search word so as to enable a first clustering result obtained by clustering to correspond to the label information.
According to an embodiment of the disclosure, the first clustering result includes a plurality of first categories, each of the plurality of first categories characterizing at least one first item. The method for processing the target search term by using the target model to obtain the target object associated with the target search term comprises the following steps: clustering the target search words by using a target model to determine target categories of the target search words in a plurality of first categories; and
determining at least one first item characterized by the target category as a target item.
According to an embodiment of the present disclosure, the apparatus 300 may further include, for example: a third obtaining module and a clustering module. And the third acquisition module acquires the model to be trained. And the clustering module is used for clustering the second historical search words by using the model to be trained to obtain an initial model, wherein a second clustering result obtained by clustering comprises a plurality of second categories, each second category in the second categories comprises at least one second historical search word, and for each category in the second categories, second articles associated with each second historical search word in the at least one second historical search word are the same.
According to an embodiment of the present disclosure, obtaining a plurality of first historical search terms includes: determining the order placing time of each historical item purchased by a user, acquiring a plurality of third historical search terms in a preset time period before the order placing time based on the order placing time for each historical item, and determining at least one third historical search term associated with the historical item in the plurality of third historical search terms as a first historical search term.
According to an embodiment of the present disclosure, the apparatus 300 may further include, for example: the association module is used for associating each first historical search word in the plurality of first historical search words with the historical item corresponding to the first historical search word so that each first historical search word has label information.
According to an embodiment of the present disclosure, processing a target search term using a target model includes: and determining whether the associated articles related to the target search term can be obtained by matching in a matching mode, and processing the target search term by using the target model in response to determining that the associated articles related to the target search term cannot be matched in the matching mode.
Any number of modules, sub-modules, units, sub-units, or at least part of the functionality of any number thereof according to embodiments of the present disclosure may be implemented in one module. Any one or more of the modules, sub-modules, units, and sub-units according to the embodiments of the present disclosure may be implemented by being split into a plurality of modules. Any one or more of the modules, sub-modules, units, sub-units according to embodiments of the present disclosure may be implemented at least in part as a hardware circuit, such as a Field Programmable Gate Array (FPGA), a Programmable Logic Array (PLA), a system on a chip, a system on a substrate, a system on a package, an Application Specific Integrated Circuit (ASIC), or may be implemented in any other reasonable manner of hardware or firmware by integrating or packaging a circuit, or in any one of or a suitable combination of software, hardware, and firmware implementations. Alternatively, one or more of the modules, sub-modules, units, sub-units according to embodiments of the disclosure may be at least partially implemented as a computer program module, which when executed may perform the corresponding functions.
For example, any of the first obtaining module 310, the processing module 320, the second obtaining module 330, and the training module 340 may be combined and implemented in one module, or any one of them may be split into multiple modules. Alternatively, at least part of the functionality of one or more of these modules may be combined with at least part of the functionality of the other modules and implemented in one module. According to an embodiment of the present disclosure, at least one of the first obtaining module 310, the processing module 320, the second obtaining module 330, and the training module 340 may be implemented at least partially as a hardware circuit, such as a Field Programmable Gate Array (FPGA), a Programmable Logic Array (PLA), a system on a chip, a system on a substrate, a system on a package, an Application Specific Integrated Circuit (ASIC), or may be implemented in hardware or firmware by any other reasonable manner of integrating or packaging a circuit, or may be implemented in any one of three implementations of software, hardware, and firmware, or in a suitable combination of any of them. Alternatively, at least one of the first acquisition module 310, the processing module 320, the second acquisition module 330 and the training module 340 may be at least partially implemented as a computer program module, which when executed may perform the respective functions.
FIG. 4 schematically illustrates a block diagram of a computer system adapted for item recommendation, in accordance with an embodiment of the present disclosure. The computer system illustrated in FIG. 4 is only one example and should not impose any limitations on the scope of use or functionality of embodiments of the disclosure.
As shown in fig. 4, a computer system 400 according to an embodiment of the present disclosure includes a processor 401 that can perform various appropriate actions and processes according to a program stored in a Read Only Memory (ROM)402 or a program loaded from a storage section 406 into a Random Access Memory (RAM) 403. Processor 401 may include, for example, a general purpose microprocessor (e.g., a CPU), an instruction set processor and/or associated chipset, and/or a special purpose microprocessor (e.g., an Application Specific Integrated Circuit (ASIC)), among others. The processor 401 may also include onboard memory for caching purposes. Processor 401 may include a single processing unit or multiple processing units for performing the different actions of the method flows in accordance with embodiments of the present disclosure.
In the RAM 403, various programs and data necessary for the operation of the system 400 are stored. The processor 401, ROM 402 and RAM 403 are connected to each other by a bus 404. The processor 401 performs various operations of the method flows according to the embodiments of the present disclosure by executing programs in the ROM 402 and/or the RAM 403. Note that the program may also be stored in one or more memories other than the ROM 402 and the RAM 403. The processor 401 may also perform various operations of the method flows according to embodiments of the present disclosure by executing programs stored in the one or more memories.
According to an embodiment of the present disclosure, system 400 may also include an input/output (I/O) interface 405, input/output (I/O) interface 405 also connected to bus 404. The system 400 may also include one or more of the following components connected to the I/O interface 405: an input section 406 including a keyboard, a mouse, and the like; an output section 407 including a display device such as a Cathode Ray Tube (CRT), a Liquid Crystal Display (LCD), and the like, and a speaker; a storage section 408 including a hard disk and the like; and a communication section 409 including a network interface card such as a LAN card, a modem, or the like. The communication section 409 performs communication processing via a network such as the internet. A driver 410 is also connected to the I/O interface 405 as needed. A removable medium 411 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, or the like is mounted on the drive 410 as necessary, so that a computer program read out therefrom is mounted into the storage section 408 as necessary.
According to embodiments of the present disclosure, method flows according to embodiments of the present disclosure may be implemented as computer software programs. For example, embodiments of the present disclosure include a computer program product comprising a computer program embodied on a computer readable storage medium, the computer program containing program code for performing the method illustrated by the flow chart. In such an embodiment, the computer program may be downloaded and installed from a network through the communication section 409, and/or installed from the removable medium 411. The computer program, when executed by the processor 401, performs the above-described functions defined in the system of the embodiments of the present disclosure. The systems, devices, apparatuses, modules, units, etc. described above may be implemented by computer program modules according to embodiments of the present disclosure.
The present disclosure also provides a computer-readable storage medium, which may be contained in the apparatus/device/system described in the above embodiments; or may exist separately and not be assembled into the device/apparatus/system. The computer-readable storage medium carries one or more programs which, when executed, implement the method according to an embodiment of the disclosure.
According to embodiments of the present disclosure, the computer-readable storage medium may be a computer-non-volatile computer-readable storage medium, which may include, for example and without limitation: a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the present disclosure, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device.
For example, according to embodiments of the present disclosure, a computer-readable storage medium may include ROM 402 and/or RAM 403 and/or one or more memories other than ROM 402 and RAM 403 described above.
The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, 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. It will also be noted that each block of the block diagrams or flowchart illustration, and combinations of blocks in the block diagrams or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
Those skilled in the art will appreciate that various combinations and/or combinations of features recited in the various embodiments and/or claims of the present disclosure can be made, even if such combinations or combinations are not expressly recited in the present disclosure. In particular, various combinations and/or combinations of the features recited in the various embodiments and/or claims of the present disclosure may be made without departing from the spirit or teaching of the present disclosure. All such combinations and/or associations are within the scope of the present disclosure.
The embodiments of the present disclosure have been described above. However, these examples are for illustrative purposes only and are not intended to limit the scope of the present disclosure. Although the embodiments are described separately above, this does not mean that the measures in the embodiments cannot be used in advantageous combination. The scope of the disclosure is defined by the appended claims and equivalents thereof. Various alternatives and modifications can be devised by those skilled in the art without departing from the scope of the present disclosure, and such alternatives and modifications are intended to be within the scope of the present disclosure.
Claims (10)
1. An item recommendation method comprising:
acquiring a target search word of a user; and
processing the target search terms by using a target model to obtain target articles associated with the target search terms, and recommending the target articles to the user;
wherein the target model is obtained based on:
obtaining a plurality of first historical search terms, wherein each of the plurality of first historical search terms has tag information, and the tag information represents historical items associated with the first historical search terms; and
and training an initial model by using the plurality of first historical search words to obtain the target model, wherein the initial model is obtained by clustering a plurality of second historical search words.
2. The method of claim 1, wherein the training an initial model using the plurality of first historical search terms comprises:
clustering each first historical search word in the plurality of first historical search words by using the initial model; and
when the initial model is used for clustering each first historical search word in the plurality of first historical search words, model parameters of the initial model are adjusted based on label information of each first historical search word, so that a first clustering result obtained through clustering corresponds to the label information.
3. The method of claim 2, wherein the first clustering result includes a plurality of first categories, each of the plurality of first categories characterizing at least one first item;
wherein the processing the target search term by using the target model to obtain the target item associated with the target search term comprises:
clustering the target search terms by using the target model to determine a target category to which the target search terms belong in the plurality of first categories; and
determining that the at least one first item characterized by the target category is the target item.
4. The method of claim 1, further comprising:
obtaining a model to be trained; and
clustering the plurality of second historical search terms by using the model to be trained to obtain the initial model,
and for each category in the plurality of second categories, the second item associated with each second historical search word in the at least one second historical search word is the same.
5. The method of any of claims 1-4, wherein the obtaining a plurality of first historical search terms comprises:
determining a placing time for a user to purchase each historical item in a plurality of historical items;
acquiring a plurality of third history search words in a preset time period before the order placing time based on the order placing time for each historical item; and
determining at least one third history search term of the plurality of third history search terms associated with the historical item as the first history search term.
6. The method of claim 1, further comprising:
and associating each first historical search word in the plurality of first historical search words with the historical item corresponding to each first historical search word so that each first historical search word has the label information.
7. The method of claim 1, wherein the processing the target search term using a target model comprises:
determining whether the related articles related to the target search terms can be obtained by matching in a matching mode; and
and in response to determining that the associated item related to the target search term is not matched by using a matching mode, processing the target search term by using the target model.
8. An item recommendation device comprising:
the first acquisition module is used for acquiring a target search word of a user; and
the processing module is used for processing the target search terms by utilizing a target model to obtain target articles related to the target search terms and recommending the target articles to the user;
wherein the target model is obtained based on:
the second acquisition module is used for acquiring a plurality of first historical search terms, wherein each first historical search term in the plurality of first historical search terms respectively has label information, and the label information represents a historical item associated with the first historical search term; and
and the training module is used for training an initial model by using the plurality of first historical search words to obtain the target model, wherein the initial model is obtained by clustering a plurality of second historical search words.
9. A computing device, comprising:
one or more processors;
a memory for storing one or more programs,
wherein the one or more programs, when executed by the one or more processors, cause the one or more processors to implement the method of any of claims 1-7.
10. A computer-readable storage medium storing computer-executable instructions for implementing the method of any one of claims 1 to 7 when executed.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010471229.9A CN112330382B (en) | 2020-05-28 | 2020-05-28 | Item recommendation method, device, computing equipment and medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010471229.9A CN112330382B (en) | 2020-05-28 | 2020-05-28 | Item recommendation method, device, computing equipment and medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112330382A true CN112330382A (en) | 2021-02-05 |
| CN112330382B CN112330382B (en) | 2024-06-18 |
Family
ID=74302904
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010471229.9A Active CN112330382B (en) | 2020-05-28 | 2020-05-28 | Item recommendation method, device, computing equipment and medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112330382B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113065932A (en) * | 2021-05-06 | 2021-07-02 | 北京京东振世信息技术有限公司 | Article recommendation method and device |
| CN113111268A (en) * | 2021-04-30 | 2021-07-13 | 百度在线网络技术（北京）有限公司 | Training method of user feature extraction model, content recommendation method and device |
| CN113240465A (en) * | 2021-05-11 | 2021-08-10 | 北京沃东天骏信息技术有限公司 | Information generation method and device |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102629257A (en) * | 2012-02-29 | 2012-08-08 | 南京大学 | Commodity recommending method of e-commerce website based on keywords |
| US20170132318A1 (en) * | 2015-11-06 | 2017-05-11 | Alibaba Group Holding Limited | Method, system, and device for item search |
| CN107194769A (en) * | 2017-05-17 | 2017-09-22 | 东莞市华睿电子科技有限公司 | A kind of Method of Commodity Recommendation that content is searched for based on user |
| CN108664513A (en) * | 2017-03-31 | 2018-10-16 | 北京京东尚科信息技术有限公司 | Method, apparatus and equipment for pushing keyword |
| CN108805622A (en) * | 2018-06-11 | 2018-11-13 | 深圳乐信软件技术有限公司 | Method of Commodity Recommendation, device, equipment and storage medium |
| CN108960992A (en) * | 2018-06-29 | 2018-12-07 | 黄宗治 | A kind of information recommendation method and relevant device |
| CN109727088A (en) * | 2018-11-08 | 2019-05-07 | 深圳壹账通智能科技有限公司 | A kind of Method of Commodity Recommendation and device based on browser |
| CN110188274A (en) * | 2019-05-30 | 2019-08-30 | 口口相传(北京)网络技术有限公司 | Search for error correction method and device |
-
2020
- 2020-05-28 CN CN202010471229.9A patent/CN112330382B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102629257A (en) * | 2012-02-29 | 2012-08-08 | 南京大学 | Commodity recommending method of e-commerce website based on keywords |
| US20170132318A1 (en) * | 2015-11-06 | 2017-05-11 | Alibaba Group Holding Limited | Method, system, and device for item search |
| CN108664513A (en) * | 2017-03-31 | 2018-10-16 | 北京京东尚科信息技术有限公司 | Method, apparatus and equipment for pushing keyword |
| CN107194769A (en) * | 2017-05-17 | 2017-09-22 | 东莞市华睿电子科技有限公司 | A kind of Method of Commodity Recommendation that content is searched for based on user |
| CN108805622A (en) * | 2018-06-11 | 2018-11-13 | 深圳乐信软件技术有限公司 | Method of Commodity Recommendation, device, equipment and storage medium |
| CN108960992A (en) * | 2018-06-29 | 2018-12-07 | 黄宗治 | A kind of information recommendation method and relevant device |
| CN109727088A (en) * | 2018-11-08 | 2019-05-07 | 深圳壹账通智能科技有限公司 | A kind of Method of Commodity Recommendation and device based on browser |
| CN110188274A (en) * | 2019-05-30 | 2019-08-30 | 口口相传(北京)网络技术有限公司 | Search for error correction method and device |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113111268A (en) * | 2021-04-30 | 2021-07-13 | 百度在线网络技术（北京）有限公司 | Training method of user feature extraction model, content recommendation method and device |
| CN113111268B (en) * | 2021-04-30 | 2024-06-11 | 百度在线网络技术（北京）有限公司 | Training method of user feature extraction model, content recommendation method and device |
| CN113065932A (en) * | 2021-05-06 | 2021-07-02 | 北京京东振世信息技术有限公司 | Article recommendation method and device |
| CN113240465A (en) * | 2021-05-11 | 2021-08-10 | 北京沃东天骏信息技术有限公司 | Information generation method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112330382B (en) | 2024-06-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107172151B (en) | Method and device for pushing information | |
| US20200250732A1 (en) | Method and apparatus for use in determining tags of interest to user | |
| US11310559B2 (en) | Method and apparatus for recommending video | |
| CN107577807B (en) | Method and device for pushing information | |
| CN112330382B (en) | Item recommendation method, device, computing equipment and medium | |
| CN110020162B (en) | User identification method and device | |
| CN108280200B (en) | Method and device for pushing information | |
| US20180101617A1 (en) | Ranking Search Results using Machine Learning Based Models | |
| CN110059172B (en) | Method and device for recommending answers based on natural language understanding | |
| CN112819512B (en) | Text processing method, device, equipment and medium | |
| CN111768258B (en) | Method, device, electronic equipment and medium for identifying abnormal order | |
| CN112650942A (en) | Product recommendation method, device, computer system and computer-readable storage medium | |
| CN107291774B (en) | Error sample identification method and device | |
| CN110827101B (en) | Shop recommending method and device | |
| CN112836128A (en) | Information recommendation method, device, equipment and storage medium | |
| US10474688B2 (en) | System and method to recommend a bundle of items based on item/user tagging and co-install graph | |
| CN112784861B (en) | Similarity determination method, device, electronic equipment and storage medium | |
| US20160124580A1 (en) | Method and system for providing content with a user interface | |
| CN113159877B (en) | Data processing method, device, system and computer readable storage medium | |
| US20130232172A1 (en) | Methods and systems for matching expressions | |
| CN114510562A (en) | Method for constructing item association graph, item query method, device and equipment | |
| CN113918810A (en) | Information pushing method, device, equipment and medium based on machine learning model | |
| CN113743973A (en) | Method and device for analyzing market hotspot trend | |
| CN111339124A (en) | Data display method and device, electronic equipment and computer readable medium | |
| US20180052874A1 (en) | Automatic content tagging |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |