CN111737418A - Method, apparatus and storage medium for predicting relevance of search term and commodity - Google Patents
Method, apparatus and storage medium for predicting relevance of search term and commodity Download PDFInfo
- Publication number
- CN111737418A CN111737418A CN202010699655.8A CN202010699655A CN111737418A CN 111737418 A CN111737418 A CN 111737418A CN 202010699655 A CN202010699655 A CN 202010699655A CN 111737418 A CN111737418 A CN 111737418A
- Authority
- CN
- China
- Prior art keywords
- search
- training
- commodity
- user
- semantic vector
- 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 50
- 239000013598 vector Substances 0.000 claims abstract description 92
- 238000012549 training Methods 0.000 claims description 171
- 230000000875 corresponding effect Effects 0.000 claims description 56
- 230000006399 behavior Effects 0.000 claims description 26
- 230000009471 action Effects 0.000 claims description 25
- 230000004044 response Effects 0.000 claims description 24
- 230000008569 process Effects 0.000 claims description 13
- 238000004590 computer program Methods 0.000 claims description 10
- 238000003062 neural network model Methods 0.000 claims description 8
- 238000013507 mapping Methods 0.000 claims description 3
- 230000002596 correlated effect Effects 0.000 claims description 2
- 238000010276 construction Methods 0.000 claims 1
- 230000014509 gene expression Effects 0.000 abstract description 14
- 235000013336 milk Nutrition 0.000 description 22
- 239000008267 milk Substances 0.000 description 22
- 210000004080 milk Anatomy 0.000 description 22
- 230000004927 fusion Effects 0.000 description 11
- 230000000694 effects Effects 0.000 description 5
- 238000012545 processing Methods 0.000 description 5
- 238000004891 communication Methods 0.000 description 4
- 230000006870 function Effects 0.000 description 4
- 235000019580 granularity Nutrition 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 235000013311 vegetables Nutrition 0.000 description 3
- UCHDWCPVSPXUMX-TZIWLTJVSA-N Montelukast Chemical compound CC(C)(O)C1=CC=CC=C1CC[C@H](C=1C=C(\C=C\C=2N=C3C=C(Cl)C=CC3=CC=2)C=CC=1)SCC1(CC(O)=O)CC1 UCHDWCPVSPXUMX-TZIWLTJVSA-N 0.000 description 2
- 240000007594 Oryza sativa Species 0.000 description 2
- 235000007164 Oryza sativa Nutrition 0.000 description 2
- 235000013305 food Nutrition 0.000 description 2
- 229960005127 montelukast Drugs 0.000 description 2
- 235000009566 rice Nutrition 0.000 description 2
- 239000004065 semiconductor Substances 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 244000300264 Spinacia oleracea Species 0.000 description 1
- 235000009337 Spinacia oleracea Nutrition 0.000 description 1
- 230000002776 aggregation Effects 0.000 description 1
- 238000004220 aggregation Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 235000008429 bread Nutrition 0.000 description 1
- 235000021152 breakfast Nutrition 0.000 description 1
- 235000020965 cold beverage Nutrition 0.000 description 1
- 235000019543 dairy drink Nutrition 0.000 description 1
- 235000013365 dairy product Nutrition 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 235000021185 dessert Nutrition 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 229910052900 illite Inorganic materials 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- VGIBGUSAECPPNB-UHFFFAOYSA-L nonaaluminum;magnesium;tripotassium;1,3-dioxido-2,4,5-trioxa-1,3-disilabicyclo[1.1.1]pentane;iron(2+);oxygen(2-);fluoride;hydroxide Chemical compound [OH-].[O-2].[O-2].[O-2].[O-2].[O-2].[F-].[Mg+2].[Al+3].[Al+3].[Al+3].[Al+3].[Al+3].[Al+3].[Al+3].[Al+3].[Al+3].[K+].[K+].[K+].[Fe+2].O1[Si]2([O-])O[Si]1([O-])O2.O1[Si]2([O-])O[Si]1([O-])O2.O1[Si]2([O-])O[Si]1([O-])O2.O1[Si]2([O-])O[Si]1([O-])O2.O1[Si]2([O-])O[Si]1([O-])O2.O1[Si]2([O-])O[Si]1([O-])O2.O1[Si]2([O-])O[Si]1([O-])O2 VGIBGUSAECPPNB-UHFFFAOYSA-L 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 235000019640 taste Nutrition 0.000 description 1
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
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/334—Query execution
- G06F16/3344—Query execution using natural language analysis
-
- 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
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
- G06F16/374—Thesaurus
-
- 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
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0623—Item investigation
- G06Q30/0625—Directed, with specific intent or strategy
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Databases & Information Systems (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Artificial Intelligence (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
Embodiments of the present disclosure provide a method, an apparatus, and a storage medium for predicting relevance of a search term and a commodity, wherein the method includes: receiving a search word input by a current user, and determining a semantic vector of the search word, wherein the semantic vector is used for representing the position of the search word in a semantic vector space of a dictionary; determining at least one search result item corresponding to the semantic vector of the search word by using a pre-trained search recommendation model, wherein the search result item is a semantic vector comprising the search word and the trade name and the category name of which the correlation with the search word is greater than a preset threshold value; determining the position of the search result item in a semantic vector space of a dictionary, and determining corresponding commodity information; and recommending the commodity information to the current user. In this way, through supervised learning, the vector expression of the word can be better learned, so that the predicted correlation result is more accurate.
Description
Technical Field
Embodiments of the present disclosure relate generally to the field of search technology, and more particularly, to a method, apparatus, and storage medium for predicting relevance of search terms and goods.
Background
The current main technology is to select all trade names as linguistic data, and use a shallow neural network to learn the vector expression of each word from the context of the trade names to express the semantics of the words. On the basis of the above, vector expressions of the search term and the product name are calculated as semantic expressions by an aggregation method such as averaging or summing. The cosine distance of the two vectors is then calculated as the correlation score.
However, the effect of the prior art depends on the richness of the corpus, the commodity title is generally short in the e-commerce field, the short text scene contains less context information, the vector expression of the word granularity is difficult to learn in the scene of limited text corpus, the semantics of the search word and the commodity name cannot be effectively expressed, and the accuracy of the recommended search result item is low when the user searches the intention commodity through the search word.
Disclosure of Invention
In view of this, according to the embodiments of the present disclosure, a scheme for predicting relevance of search terms and goods is provided, which satisfies the requirement of improving accuracy of search result items, and further improves user experience.
In a first aspect of the present disclosure, there is provided a method for predicting relevance of a search term to a commodity, including:
receiving a search word input by a current user, and determining a semantic vector of the search word, wherein the semantic vector is used for representing the position of the search word in a semantic vector space of a dictionary;
determining at least one search result item corresponding to the semantic vector of the search word by using a pre-trained search recommendation model, wherein the search result item is a semantic vector comprising the search word and the trade name and the category name of which the correlation with the search word is greater than a preset threshold value;
determining the position of the search result item in a semantic vector space of a dictionary, and determining corresponding commodity information;
and recommending the commodity information to the current user.
As for the above-mentioned aspect and any possible implementation manner, an implementation manner is further provided, which further includes a process of building a semantic vector space of a dictionary, specifically including:
acquiring search terms, trade names and category names in a historical search behavior log of a user;
and carrying out line-by-line coding on the search words, the trade names and the class names to generate a semantic vector space, wherein the codes of the search words, the trade names and the class names are respectively corresponding line numbers.
The above-described aspects and any possible implementation further provide an implementation, and the search recommendation model is trained by:
acquiring a behavior log of a user in a preset time period, wherein the behavior log comprises a search request of the user and a response action of the user to commodity information returned according to the search request;
taking the search word corresponding to the commodity information with the click rate higher than the preset threshold value and the commodity names and the class names in the commodity information as training positive samples;
taking the search word corresponding to the commodity information with the click rate lower than the preset threshold value and the commodity names and the class names in the commodity information as first sub-training negative samples;
selecting a search word corresponding to the commodity information and a commodity name and a class name in the commodity information from the class to which the training positive sample belongs and a superior class as second sub-training negative samples according to preset conditions;
fusing the first sub-training negative sample and the second sub-training negative sample to generate a training negative sample;
adding class characteristics into the training positive sample and the training negative sample in a correlated manner, generating a characteristic training positive sample and a characteristic training negative sample, and mapping the characteristic training positive sample and the characteristic training negative sample into semantic vectors in a semantic vector space;
and training a neural network model by using the semantic vectors of the feature training positive sample and the feature training negative sample to generate a search recommendation model.
The above-described aspects and any possible implementation further provide an implementation, further including:
selecting search words corresponding to commodity information in a preset proportion and commodity names and category names in the commodity information from different categories to which the training positive sample belongs as a third sub-training negative sample;
the fusing the first sub-training negative sample and the second sub-training negative sample to generate a training negative sample, including:
and fusing the first sub-training negative sample, the second sub-training negative sample and the third sub-training negative sample to generate a training negative sample.
The above-described aspects and any possible implementation further provide an implementation, where the search recommendation model is one of an FM model, an FFM model, a bilinear FFM model, and a depeffm model.
The above-described aspect and any possible implementation further provide an implementation, where the class names include multiple levels of class names, and each level of class names corresponds to one dimension of the semantic vector space.
The above-described aspect and any possible implementation further provide an implementation, where the semantic vector space further includes a user information dimension, and the user information includes at least one of a user ID, a user gender, and a user occupation.
The above aspect and any possible implementation manner further provide an implementation manner, where the receiving a search term input by a current user, and determining a semantic vector of the search term includes:
receiving a search word input by a current user, acquiring user information of the current user, and determining the search word and a semantic vector of the user information.
In a second aspect of the disclosure, an electronic device is provided, comprising a memory having stored thereon a computer program and a processor implementing the method as described above when executing the program.
In a third aspect of the disclosure, a computer-readable storage medium is provided, on which a computer program is stored, which program, when being executed by a processor, is adapted to carry out the method as set forth above.
It should be understood that the statements herein reciting aspects are not intended to limit the critical or essential features of the embodiments of the present disclosure, nor are they intended to limit the scope of the present disclosure. Other features of the present disclosure will become apparent from the following description.
According to the relevance prediction scheme of the search terms and the commodities, positive and negative samples are directly constructed for the search terms, the commodity names and the categories in the training stage, and vector expression of the terms can be better learned through supervised learning, so that the predicted relevance result is more accurate.
Drawings
The above and other features, advantages and aspects of various embodiments of the present disclosure will become more apparent by referring to the following detailed description when taken in conjunction with the accompanying drawings. In the drawings, like or similar reference characters designate like or similar elements, and wherein:
fig. 1 is a flowchart illustrating a method for predicting relevance of a search term and a commodity according to a first embodiment of the present disclosure;
FIG. 2 is a flowchart of a search recommendation model generation method according to a second embodiment of the disclosure;
fig. 3 is a schematic structural diagram illustrating a correlation prediction apparatus for a search term and a product according to a third embodiment of the present disclosure.
Detailed Description
To make the objects, technical solutions and advantages of the embodiments of the present disclosure more clear, the technical solutions of the embodiments of the present disclosure will be described clearly and completely with reference to the drawings in the embodiments of the present disclosure, and it is obvious that the described embodiments are some, but not all embodiments of the present disclosure. All other embodiments, which can be derived by a person skilled in the art from the embodiments disclosed herein without making any creative effort, shall fall within the protection scope of the present disclosure.
In addition, the term "and/or" herein is only one kind of association relationship describing an associated object, and means that there may be three kinds of relationships, for example, a and/or B, which may mean: a exists alone, A and B exist simultaneously, and B exists alone. In addition, the character "/" herein generally indicates that the former and latter related objects are in an "or" relationship.
In the present disclosure, semantic expression of words is learned based on the association relationship between search words and product names in a search behavior log. In the training stage, positive and negative samples are directly constructed for the search words, trade names and categories, and vector expression of the words can be better learned through supervised learning. Meanwhile, the category characteristics are introduced into the trained samples, and category information can be integrated into vector expression of words.
Specifically, as shown in fig. 1, it is a flowchart of a method for predicting relevance between a search term and a product according to a first embodiment of the present disclosure. As shown in fig. 1, the method of this embodiment may include the following steps:
s101: receiving a search word input by a current user, and determining a semantic vector of the search word, wherein the semantic vector is used for representing the position of the search word in a semantic vector space of a dictionary.
The method for predicting the relevance between the search terms and the commodities can be applied to Internet commodity sales APP. The APP can be provided with a search box, and a user can input search words in the search box and click a search button to send a search request. After receiving the search request, the back-end server matches the search terms in the search request from the commodity document library according to the keywords, determines commodity documents related to the keywords, further determines commodities related to the keywords as search result items, and feeds the search result items back to the user. For example, if the search word input by the user is "milk", the fed-back commodity information may be information of commodities such as "milk 250ml by 12 boxes of Mongolian Telun milk", "Ehrlich pure milk gift box 250ml by 12", or "Mongolian Lele milk 240 ml". The user can click and browse the fed back commodity information, and can further join the shopping cart or pay for purchase.
Specifically, after receiving a search word input by a user, a background server of the APP needs to determine a semantic vector of the search word, where the semantic vector is used to represent a position of the search word in a semantic vector space of a dictionary. Wherein, the dictionary can be obtained by the following method:
acquiring search terms, trade names and category names in a historical search behavior log of a user; and coding the search word, the trade name and the class name to generate a three-dimensional semantic vector space, wherein the codes of the search word, the trade name and the class name are respectively corresponding line numbers.
Furthermore, in some embodiments of the present disclosure, the class names may be divided into multiple levels, such as a primary class name, a secondary class name, and the like, and the class name of each level may be used as one dimension of the semantic vector space. Furthermore, the user ID, the user gender, the age, the occupation, etc. can also be used as one dimension of the semantic vector space. Therefore, a plurality of semantic vector spaces are formed, and the embodiment takes a three-dimensional vector space as an example to exemplarily illustrate the technical scheme of the disclosure, but should not be construed as a limitation to the technical scheme of the disclosure.
For example, in the history search behavior log, the search word input by the user is "milk", and the user clicks/adds/purchases "montelukast milk 250ml × 12 boxes", the behavior log of the user includes fields of "milk" and "montelukast milk 250ml × 12 boxes", where "delusterask" is a trade name, "250 ml × 12 boxes" is a product name, and "milk" is a search word, so that the search word, the trade name, and the product name in all the search behavior logs can be extracted in a similar manner. And then encoding the extracted search words, trade names and item names so that each search word, trade name and item name corresponds to a character string which can be a line number or a code generated according to a commodity item classification table, thereby generating a three-dimensional semantic vector space. In the three-dimensional semantic vector space, each combination of the search term, the commodity name and the category name uniquely corresponds to one commodity information.
After receiving a search word input by a user, converting the search word into a semantic vector with the dimensions of a trade name and a category name being 0, and outputting the semantic vector comprising the search word and the trade name and the category name which have correlation with the search word and meet preset conditions through a pre-trained search recommendation model. After determining at least one search result item corresponding to the semantic vector of the search word by using the search recommendation model, calculating a correlation between the semantic vector of the search word and the semantic vector of each search result item, in this embodiment, a cosine distance value between the semantic vector of the search word and the semantic vector of each search result item may be used as a correlation value between the semantic vector of the search word and the semantic vector of each search result item. And generally, it is considered that the larger the cosine distance value is, the closer the corresponding search word and the search result item are.
S102: and determining at least one search result item corresponding to the semantic vector of the search word by using a pre-trained search recommendation model, wherein the search result item is a semantic vector comprising the search word and the name and category name of the trade name and category name of which the correlation with the search word is greater than a preset threshold value.
Because the effect of the prior art depends on the richness of the linguistic data, the commodity titles in the E-commerce field are generally short and are short text scenes, the contained context information is less, and the vector expression of the word granularity is difficult to learn under the scene of limited text linguistic data, so that the semantics of search words and trade names cannot be effectively expressed, and the accuracy of the recommended search result items is lower when a user searches for the intention commodity through the search words. There is a need to enrich the corpus to make the recommended search result items more accurate. Therefore, when training the search recommendation model, a rich corpus sample needs to be selected to train the neural network model. For a specific training process of the search recommendation model, refer to the second embodiment, which is not described in detail. When the search recommendation model is trained, the selected sample is a semantic vector of a search result item in a behavior log of a user in an APP, and therefore the semantic vector of the search result item is also output through the search recommendation model.
The search result item is a semantic vector including the search word and a trade name and a category name of which the correlation with the search word is greater than a preset threshold. For example, if the user inputs "milk", the output is a semantic vector corresponding to "milk 250ml by 12 boxes", "illite classic pure milk gift box 250ml by 12", or "milk 240ml by 240ml of montreal le pillow pure milk", and the correlation between the trade name and the category name in the output result and the input keyword is greater than a preset threshold. It should be noted that the search recommendation model of the present embodiment is mainly used for determining the trade names and category names whose correlation with the search term is greater than a preset threshold.
S103: and determining the position of the search result item in a semantic vector space of a dictionary, and determining a corresponding commodity information search word.
After the semantic vector of the search result item is output through the search recommendation model, the position of the search result item in the semantic vector space of the dictionary is determined according to the semantic vector, and then corresponding commodity information is determined according to the position.
S104: and recommending the commodity information to the current user.
In some embodiments, the semantic vector space further comprises a user information dimension, the user information comprising at least one of a user ID, a user gender, and a user occupation. After receiving the search word input by the current user, the method can further acquire the user information of the current user and determine the semantic vector of the search word and the user information. And inputting the determined semantic vectors of the search terms and the user information into the search recommendation model, and determining at least one search result item corresponding to the search terms and the semantic vectors of the user information.
According to the relevance prediction scheme of the search terms and the commodities, positive and negative samples are directly constructed for the search terms, the commodity names and the class names in the training stage, and vector expression of the terms can be better learned through supervised learning, so that the predicted relevance result is more accurate.
Fig. 2 is a flowchart of a search recommendation model generation method according to a second embodiment of the present disclosure. In this embodiment, the search recommendation model may be generated by:
s201: and acquiring a behavior log of the user in a preset time period, wherein the behavior log comprises a search request of the user and a response action of the user to commodity information returned according to the search request.
In the process of generating a model by using a currently adopted negative sample strategy, a commodity exposed in one search session but not clicked/picked up/purchased is generally used as a negative sample, and a commodity clicked/picked up/purchased is used as a positive sample. Most of the search results are relevant, and whether the user clicks/adds/purchases the product is also influenced by various factors such as price and preference, and is not completely dependent on the relevance. Only a few goods that are not clicked/picked/purchased are truly irrelevant. The negative sample strategy cannot collect enough effective irrelevant samples, and part of relevant commodities appear in the negative samples to form the effect that noise data influence the relevance model.
The behavior log-based search recommendation model generation method is used for acquiring model training samples according to search requests in historical behavior logs of users and response actions of the users to commodity information returned according to the search requests, further training the search recommendation model, generating the search recommendation model, and recommending commodities for the users according to the search requests of the users by using the generated search recommendation model, so that recommendation results are more accurate.
Specifically, the technical solution of the present application is described by taking an e-commerce APP as an example. The APP can be provided with a search box, and a user can input related information of commodities in the search box and click a search button to send a search request. And after receiving the search request, the back-end server matches the search request from the commodity document library according to the keywords in the commodity related information in the search request, determines the commodity documents related to the keywords, further determines the commodities related to the keywords, and feeds the information of the commodities back to the user. The user can click and browse the fed back commodity information, and can further join the shopping cart or pay for purchase. The operation of the user on the feedback information is a response action. And, the search request of the user and the response action of the user to the commodity information returned according to the search request are subjected to data storage in a behavior log mode through the APP. When the search recommendation model is generated by using the method of the embodiment, the behavior log of the user within the preset time period can be acquired.
In some embodiments, the preset time period may be two weeks, or the specific duration of the preset time period may also be determined according to actual needs, such as one month, one quarter, half a year, and the like.
S202: and taking the search word corresponding to the commodity information with the click rate higher than the preset threshold value and the commodity name and the class name in the commodity information as training positive samples.
In this embodiment, the user may click and browse the fed back commodity information, may further join the shopping cart, or may pay for purchase again, and of course, if the fed back commodity information is not the commodity information expected by the user, the user may not perform any operation on the fed back commodity information, for example, may re-input new commodity information in the search box, for example, re-search by changing "vegetables" to "fruits". Also, the product information fed back for similar search information may be the same. However, the response actions of different users are different for the same feedback information. In this embodiment, the response actions are divided into an un-click response action and a click response action, wherein the click response action can be further divided into click browsing, click to add a cart (click to add a shopping cart), and click to purchase. And counting the behavior logs of the user in a preset time period, and taking the search word corresponding to the commodity information with the click rate higher than a preset threshold value in the commodity information and the commodity name and the category name in the commodity information as a training positive sample.
In some embodiments, the search term corresponding to the commodity information with the click through rate higher than the preset threshold value and the commodity name and the commodity category name in the commodity information are used as training positive samples, or the search term corresponding to the commodity information with the click through rate higher than the preset threshold value and the commodity name and the commodity category name in the commodity information are used as training positive samples.
In addition, in some embodiments, the search word corresponding to the commodity information with the click through rate higher than the preset threshold and the commodity name and the commodity category name in the commodity information are used as the training positive samples, the search word corresponding to the commodity information with the click through rate higher than the preset threshold and the commodity name and the commodity category name in the commodity information, the search word corresponding to the commodity information with the click through rate higher than the preset threshold and two of the commodity name and the commodity category name in the commodity information are used as the training positive samples, or the search word corresponding to the commodity information with the click through rate higher than the preset threshold and the commodity name and the commodity category name in the commodity information, the search word corresponding to the commodity information with the click through rate higher than the preset threshold and the search word corresponding to the commodity information with the commodity name, the commodity category name and the click through rate higher than the preset threshold, And selecting a search word corresponding to the commodity information and the commodity name and the class name in the commodity information according to a proportion from the commodity name and the class name in the commodity information to serve as a training positive sample.
Alternatively, the product information may be proportionally selected from the search term corresponding to the product information in which the click browsing rate is higher than the preset threshold value, the product name and the class name in the product information, the search term corresponding to the product information in which the click car-adding rate is higher than the preset threshold value, the product name and the class name in the product information and the search term corresponding to the product information in which the click purchase rate is higher than the preset threshold value, and the product name and the class name in the product information, as the training positive sample. For example, the search term corresponding to 20% of the commodity information and the commodity name and the commodity category name in the commodity information may be selected from the commodity information having the click browsing rate higher than the preset threshold value, the search term corresponding to 30% of the commodity information and the commodity name and the commodity category name in the commodity information may be selected from the commodity information having the click car-adding rate higher than the preset threshold value, and the search term corresponding to 40% of the commodity information and the commodity name and the commodity category name in the commodity information may be selected from the commodity information having the click purchase rate higher than the preset threshold value. The accuracy of model recognition is improved by improving the search terms corresponding to the commodity information with the click purchase rate higher than the preset threshold value and the ratio of the commodity names and the class names in the commodity information in the training sample.
Or, the commodity information may be selected from the commodity information in which the click browsing rate is higher than the preset threshold, the commodity information in which the click car-adding rate is higher than the preset threshold, and the commodity information in which the click purchasing rate is higher than the preset threshold, respectively, according to a preset ratio, for example, the commodity information in which the click browsing rate is higher than the preset threshold, the commodity information in which the click car-adding rate is higher than the preset threshold, and the corresponding search word in the commodity information in which the click purchasing rate is higher than the preset threshold, and the commodity name and the commodity category name in the commodity information may be selected as training positive samples according to a ratio of 6:3:1, where the total ratio of the commodity information amount in which the click purchasing rate is higher than the preset threshold, and the commodity information amount in which the click car-adding rate is higher than the preset threshold is 6:3: 1.
The method improves the accuracy of model identification by improving the proportion of commodity information with the click purchase rate higher than a preset threshold value in the training sample.
S203: and taking the search word corresponding to the commodity information with the click rate lower than the preset threshold value and the commodity name and the class name in the commodity information as a first sub-training negative sample.
Similarly, by means of statistics, search terms corresponding to the commodity information of which the click rate is lower than a preset threshold value in the commodity information, and commodity names and category names in the commodity information are taken as first sub-training negative samples.
S204: and selecting a search word corresponding to the commodity information from the class to which the training positive sample belongs and the superior class according to a preset condition, and taking the commodity name and the class name in the commodity information as a second sub-training negative sample.
Because the negative sample strategy in the prior art cannot acquire enough effective negative samples, and the recommendation result generated by the existing search recommendation model is inaccurate, the embodiment of the disclosure expands the negative samples, and can select the search word corresponding to the commodity information and the commodity name and the class name in the commodity information from the class to which the training positive sample belongs and the superior class as the second sub-training negative sample. For example, if the selected positive sample is "spinach", the category to which the positive sample belongs is "vegetable", and the upper-level category is "fresh food", the search term corresponding to other commodity information and the name of the commodity and the category in the commodity information can be selected from the "vegetable" and the "fresh food" as the second sub-training negative sample. By sampling from the commodities with low click rates in the behavior logs as negative samples, real irrelevant commodities can be extracted. The search term and the item are not verified to be irrelevant in just one search session without being clicked/picked/purchased. The noise in the negative sample data can be effectively reduced by eliminating the data. Meanwhile, commodities under the same grade/upper grade to which the positive sample commodity belongs are in certain relation with the positive sample, but contain a large number of irrelevant commodities, and the irrelevant commodities can be used as a negative sample with fine granularity.
S205: and fusing the first sub-training negative sample and the second sub-training negative sample to generate a training negative sample.
In some embodiments, the first sub-training negative example and the second sub-training negative example may be directly fused to generate the training negative example.
In some embodiments, the first sub-training negative sample and the second sub-training negative sample may be proportionally fused to generate a training negative sample.
Specifically, the corresponding search recommendation models may be respectively generated according to the training negative samples generated by direct fusion and the training negative samples generated by proportional fusion, and in the actual application process, when a search request input by a user is received, one of the two search recommendation models is randomly selected, a commodity is recommended to the user according to the received search request by using the selected search recommendation model, and a response action of the user for the recommended commodity is recorded. And in a preset time period, determining which search recommendation model has more accurate recommendation results according to response actions of the user on commodities recommended by different search recommendation models, if the recommendation results of the search recommendation models directly fused to generate the training negative sample are more accurate, re-determining the proportion of the training negative sample generated by proportional fusion, and then repeating the process to optimize the training negative sample of the search recommendation model. And if the recommendation result of the training negative sample search recommendation model generated by proportional fusion is more accurate, fusing the first sub-training negative sample and the second sub-training negative sample according to the current proportion to generate a training negative sample, and taking the search recommendation model corresponding to the training negative sample as the applied search recommendation model. In addition, in the above embodiment, the fusion ratio may be adjusted periodically according to a ratio of sample amounts of the first sub-training negative sample and the second sub-training negative sample, and the training negative sample of the search recommendation model may be optimized according to the method.
S206: and associating and adding class characteristics in the training positive sample and the training negative sample to generate a characteristic training positive sample and a characteristic training negative sample, and mapping the characteristic training positive sample and the characteristic training negative sample into a semantic vector in a semantic vector space.
After the training positive sample and the training negative sample are generated, the class characteristics of the commodity can be further added into the samples to generate the characteristic training positive sample and the characteristic training negative sample, and the characteristic training positive sample and the characteristic training negative sample are mapped into the semantic vector in the semantic vector space.
The semantic vector space in this embodiment may also be a multidimensional vector space, and may include, in addition to three dimensions, that is, search terms, trade names, and category names, category feature dimensions, which may be, for example, promotions, activities, offers, and the like, and may also be inherent features of the product itself, and further formed dimensions, such as, for example, tastes. After the feature training positive sample and the feature training negative sample are generated, the generated feature training positive sample and the feature training negative sample are mapped into a semantic vector in a semantic vector space.
S207: and training a neural network model by using the training positive sample and the training negative sample to generate a search recommendation model.
After the feature training positive sample and the feature training negative sample are generated, a search request can be used as input, the feature training positive sample and the feature training negative sample are used as output, the neural network model is trained, parameters of the neural network model are modified according to the feature training positive sample and the feature training negative sample until the difference degree between the output of the model and the feature training positive sample and the feature training negative sample is smaller than a preset threshold value, and a search recommendation model is generated.
The search recommendation model in this embodiment may be one of an FM model, an FFM model, a bilinear FFM model, and a depeffm model, which is not specifically limited herein.
According to the search recommendation model generation method based on the behavior log, the negative samples with different granularities are extracted based on the commodity category system, enough effective negative samples can be collected, noise data are effectively reduced, data quality is improved, the recommendation result generated by the existing search recommendation model is more accurate, and therefore user experience is improved.
In addition, in some embodiments, commodity information in a preset proportion may be selected from different categories to which the training positive samples belong as third sub-training negative samples, and the first sub-training negative sample, the second sub-training negative sample, and the third sub-training negative sample are fused to generate the training negative samples.
Most commodities are different from the positive samples greatly under the categories except the positive samples, and the commodities can be used as coarse-grained negative samples. The negative sample extraction strategy effectively reduces noise data and improves data quality.
In some embodiments, the samples and models may be continually updated with new historical records, such that the generated search recommendation model is adapted to the iteration and updating of the commodity.
In some embodiments, the neural network model may be further trained by using the training positive samples and the training negative samples, so as to generate corresponding positive search recommendation models and corresponding negative search recommendation models, where the positive search recommendation models are used for predicting commodity information with a high click rate, and the negative search recommendation models are used for predicting commodity information with a low click rate.
And grouping the response actions according to the relevance among the keywords of the search request, and combining the search requests corresponding to the keywords with the relevance higher than a preset threshold value into the same search request, so that the response actions are divided into a plurality of response action groups, and each response action group is associated with one search request.
The method of this embodiment learns the semantic expression of the word based on the commonalities between the search word and the product name in the search behavior log. The user search logs are large in quantity, rich and diverse, and good linguistic data are provided for relevance learning. In the training stage, positive and negative samples are directly constructed for the search words, trade names and categories, and vector expression of the words can be better learned through supervised learning. And (3) introducing category characteristics into the trained sample, so that category information can be integrated into vector expression of words. As an alternative embodiment of the present disclosure, the above embodiment may further include: and selecting commodity information with a preset proportion from different categories to which the training positive sample belongs as a third sub-training negative sample, and fusing the first sub-training negative sample, the second sub-training negative sample and the third sub-training negative sample to generate a training negative sample.
When the training negative samples are generated, the generated training negative samples can be directly fused, or the generated training negative samples can be fused in proportion, corresponding search recommendation models are respectively generated according to the training negative samples generated by direct fusion and the training negative samples generated by proportional fusion, in the practical application process, when a search request input by a user is received, one of the two search recommendation models is randomly selected, the selected search recommendation model is used for recommending commodities for the user according to the received search request, and the response action of the user for the recommended commodities is recorded. And in a preset time period, determining which search recommendation model has more accurate recommendation results according to response actions of the user on commodities recommended by different search recommendation models, if the recommendation results of the search recommendation models directly fused to generate the training negative sample are more accurate, re-determining the proportion of the training negative sample generated by proportional fusion, and then repeating the process to optimize the training negative sample of the search recommendation model. And if the recommendation result of the training negative sample search recommendation model generated by proportional fusion is more accurate, fusing the first sub-training negative sample, the second sub-training negative sample and the third sub-training sample according to the current proportion to generate a search recommendation model corresponding to the training negative sample as the applied search recommendation model.
As an optional embodiment of the present disclosure, the training negative samples generated by proportional fusion in the above embodiments may be obtained by fusing the first sub-training negative sample, the second sub-training negative sample, and the third sub-training negative sample according to two different proportional values to generate training negative samples, and respectively generate corresponding search recommendation models, in an actual application process, when a search request input by a user is received, one of the two search recommendation models is randomly selected, a commodity is recommended to the user according to the received search request by using the selected search recommendation model, and a response action of the user for the recommended commodity is recorded. And in a preset time period, determining which search recommendation model has a more accurate recommendation result according to the response action of the user on the commodities recommended by different search recommendation models, and adjusting the fusion ratio value according to the variation trend of the two ratio values. For example, two different proportional values of the first sub-training negative example, the second sub-training negative example, and the third sub-training negative example are 1: 3: 5 and 1: 4: 7, and the ratio is 1: 4: 7, if the recommendation result of the search recommendation model is more accurate, the ratio of the first sub-training negative example, the second sub-training negative example, and the third sub-training negative example may be increased, for example, the ratio may be changed to 1: 5: 9, otherwise, the ratio of the first sub-training negative example, the second sub-training negative example, and the third sub-training negative example may be reduced, for example, to 1: 2: 3. by repeating the above process, the search recommendation model can be dynamically optimized, so that the recommendation result is more accurate. The method of the present embodiment can achieve similar technical effects as those of the above embodiments, and will not be repeated herein.
In order to make the technical solution of the present disclosure more easily understood, the technical solution of the present disclosure is further described below with a specific application example. The user's historical behavior log may be search requests of a plurality of users acquired using big data technology and response actions of each user to merchandise information returned according to the search requests. The search request may include a search term, and based on the search term, an item associated with the search term and clicked for purchase in a response action of the user to item information returned according to the search request may be recommended to the user as a search result, which is mainly for different brands or different specifications of the same item, such as a brand of rice or a specification of rice (e.g., a few jin of bread), and the like.
Furthermore, as an embodiment of the present disclosure, the historical behavior log of the user may also be for the same user, i.e., the shopping history of the current user. In this embodiment, the historical behavior log of the user includes search terms and search time input by the user. For example, if the user entered milk and was in the morning, the recommended good may be breakfast milk; if the user inputs milk and it is at noon, the recommended item may be a dairy drink; the user inputs milk and in the afternoon, the recommended item may be a dairy dessert or the like. In addition, the recommended commodity can be dynamically adjusted according to time information such as seasons, for example, in winter, if the user inputs milk, the recommended commodity is hot milk, and in summer, if the user inputs milk, the recommended commodity is a cow cold drink, and the like.
In some embodiments, the neural network model may be further trained by using the training positive samples and the training negative samples, so as to generate corresponding positive search recommendation models and corresponding negative search recommendation models, where the positive search recommendation models are used for predicting commodity information with a high click rate, and the negative search recommendation models are used for predicting commodity information with a low click rate. After receiving a search word input by a current user, respectively inputting the search word into a generated positive search recommendation model and a generated negative search recommendation model, respectively outputting a positive search result and a negative search result, then determining an intersection in the positive search result and the negative search result, removing commodities contained in the intersection from the positive search result, and then recommending other search results in the positive search result to the user. In addition, the response action of the current user for the commodity in the search recommendation result can be added to the behavior log, and in the subsequent model training process, the commodity which is not clicked by the user in the search recommendation result can be used as a negative training sample to perform update training on the model.
Fig. 3 is a schematic structural diagram illustrating a correlation prediction apparatus for a search term and a product according to a third embodiment of the present disclosure. The terminal device shown in fig. 3 is only an example, and should not bring any limitation to the functions and the scope of use of the embodiments of the present application.
As shown in fig. 3, the computer system includes a Central Processing Unit (CPU) 301 that can perform various appropriate actions and processes based on a program stored in a Read Only Memory (ROM) 302 or a program loaded from a storage section 308 into a Random Access Memory (RAM) 303. In the RAM303, various programs and data necessary for system operation are also stored. The CPU 301, ROM 302, and RAM303 are connected to each other via a bus 304. An input/output (I/O) interface 305 is also connected to bus 304.
The following components are connected to the I/O interface 305: an input portion 306 including a keyboard, a mouse, and the like; an output section 307 including a display such as a Cathode Ray Tube (CRT), a Liquid Crystal Display (LCD), and the like, and a speaker; a storage section 308 including a hard disk and the like; and a communication section 309 including a network interface card such as a LAN card, a modem, or the like. The communication section 309 performs communication processing via a network such as the internet. The driver 310 is also connected to the I/O interface 305 on an as needed basis. A removable medium 311 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, or the like is mounted on the drive 310 on an as-needed basis, so that a computer program read out therefrom is mounted on the storage section 308 on an as-needed basis.
In particular, based on the embodiments of the present disclosure, the processes described above with reference to the flowcharts 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 medium, the computer program comprising program code for performing the method illustrated in the flow chart. In such an embodiment, the computer program may be downloaded and installed from a network through the communication section 309, and/or installed from the removable medium 311. The computer program performs the above-described functions defined in the method of the present application when executed by the Central Processing Unit (CPU) 301.
The functions described herein above may be performed, at least in part, by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: a Field Programmable Gate Array (FPGA), an Application Specific Integrated Circuit (ASIC), an Application Specific Standard Product (ASSP), a system on a chip (SOC), a load programmable logic device (CPLD), and the like.
Program code for implementing the methods of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus, such that the program codes, when executed by the processor or controller, cause the functions/operations specified in the flowchart and/or block diagram to be performed. The program code may execute entirely on the machine, partly on the machine, as a stand-alone software package partly on the machine and partly on a remote machine or entirely on the remote machine or server.
In the context of this disclosure, a machine-readable medium may be a tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, 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), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
Further, while operations are depicted in a particular order, this should be understood as requiring that such operations be performed in the particular order shown or in sequential order, or that all illustrated operations be performed, to achieve desirable results. Under certain circumstances, multitasking and parallel processing may be advantageous. Likewise, while several specific implementation details are included in the above discussion, these should not be construed as limitations on the scope of the disclosure. Certain features that are described in the context of separate embodiments can also be implemented in combination in a single implementation. Conversely, various features that are described in the context of a single implementation can also be implemented in multiple implementations separately or in any suitable subcombination.
Although the subject matter has been described in language specific to structural features and/or methodological acts, it is to be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or acts described above. Rather, the specific features and acts described above are disclosed as example forms of implementing the claims.
Claims (10)
1. A method for predicting the relevance of a search term to a commodity, comprising:
receiving a search word input by a current user, and determining a semantic vector of the search word, wherein the semantic vector is used for representing the position of the search word in a semantic vector space of a dictionary;
determining at least one search result item corresponding to the semantic vector of the search word by using a pre-trained search recommendation model, wherein the search result item is a semantic vector comprising the search word and the trade name and the category name of which the correlation with the search word is greater than a preset threshold value;
determining the position of the search result item in a semantic vector space of a dictionary, and determining corresponding commodity information;
and recommending the commodity information to the current user.
2. The method for predicting the relevance of search terms and commodities according to claim 1, further comprising a construction process of a semantic vector space of a dictionary, specifically comprising:
acquiring search terms, trade names and category names in a historical search behavior log of a user;
and carrying out line-by-line coding on the search words, the trade names and the class names to generate a semantic vector space, wherein the codes of the search words, the trade names and the class names are respectively corresponding line numbers.
3. The method for predicting the relevance of search terms and commodities according to claim 2, wherein the search recommendation model is trained by:
acquiring a behavior log of a user in a preset time period, wherein the behavior log comprises a search request of the user and a response action of the user to commodity information returned according to the search request;
taking the search word corresponding to the commodity information with the click rate higher than the preset threshold value and the commodity names and the class names in the commodity information as training positive samples;
taking the search word corresponding to the commodity information with the click rate lower than the preset threshold value and the commodity names and the class names in the commodity information as first sub-training negative samples;
selecting a search word corresponding to the commodity information and a commodity name and a class name in the commodity information from the class to which the training positive sample belongs and a superior class as second sub-training negative samples according to preset conditions;
fusing the first sub-training negative sample and the second sub-training negative sample to generate a training negative sample;
adding class characteristics into the training positive sample and the training negative sample in a correlated manner, generating a characteristic training positive sample and a characteristic training negative sample, and mapping the characteristic training positive sample and the characteristic training negative sample into semantic vectors in a semantic vector space;
and training a neural network model by using the semantic vectors of the feature training positive sample and the feature training negative sample to generate a search recommendation model.
4. The method of predicting the relevance of a search term to a commodity according to claim 3, further comprising:
selecting search words corresponding to commodity information in a preset proportion and commodity names and category names in the commodity information from different categories to which the training positive sample belongs as a third sub-training negative sample;
the fusing the first sub-training negative sample and the second sub-training negative sample to generate a training negative sample, including:
and fusing the first sub-training negative sample, the second sub-training negative sample and the third sub-training negative sample to generate a training negative sample.
5. The method of predicting the association between a search term and a product according to claim 4, wherein the search recommendation model is one of an FM model, an FFM model, a bilinear FFM model, and a DeepFFM model.
6. The method of claim 5, wherein the category names include a plurality of levels of category names, and each level of category names corresponds to one dimension of the semantic vector space.
7. The method of claim 6, wherein the semantic vector space further comprises a user information dimension, and the user information comprises at least one of a user ID, a user gender, and a user occupation.
8. The method for predicting the relevance of a search term and a commodity according to claim 7, wherein the receiving a search term input by a current user and determining a semantic vector of the search term comprises:
receiving a search word input by a current user, acquiring user information of the current user, and determining the search word and a semantic vector of the user information.
9. An electronic device comprising a memory and a processor, the memory having stored thereon a computer program, wherein the processor, when executing the program, implements the method of any of claims 1-8.
10. A computer-readable storage medium, on which a computer program is stored, which program, when being executed by a processor, carries out the method according to any one of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010699655.8A CN111737418B (en) | 2020-07-20 | 2020-07-20 | Method, apparatus and storage medium for predicting relevance of search term and commodity |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010699655.8A CN111737418B (en) | 2020-07-20 | 2020-07-20 | Method, apparatus and storage medium for predicting relevance of search term and commodity |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111737418A true CN111737418A (en) | 2020-10-02 |
| CN111737418B CN111737418B (en) | 2021-05-14 |
Family
ID=72655149
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010699655.8A Active CN111737418B (en) | 2020-07-20 | 2020-07-20 | Method, apparatus and storage medium for predicting relevance of search term and commodity |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111737418B (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112328889A (en) * | 2020-11-23 | 2021-02-05 | 北京字节跳动网络技术有限公司 | Method and device for determining recommended search terms, readable medium and electronic equipment |
| CN113111800A (en) * | 2021-04-19 | 2021-07-13 | 北京每日优鲜电子商务有限公司 | Commodity shelf-loading judging method and device for unmanned container and storage medium |
| KR20210151017A (en) * | 2020-11-24 | 2021-12-13 | 베이징 바이두 넷컴 사이언스 테크놀로지 컴퍼니 리미티드 | Method and apparatus for training search model, and method and apparatus for searching for target object |
| CN114139041A (en) * | 2022-01-28 | 2022-03-04 | 浙江口碑网络技术有限公司 | Category relevance prediction network training and category relevance prediction method and device |
| CN114282965A (en) * | 2021-12-21 | 2022-04-05 | 唯品会(广州)软件有限公司 | Recall scheme evaluation method and device and computer equipment |
| JP2022173084A (en) * | 2021-05-06 | 2022-11-17 | ネイバー コーポレーション | Method for searching for item based on burying similarity, computer device, and computer program |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20180268438A1 (en) * | 2012-10-04 | 2018-09-20 | Groupon, Inc. | Method, apparatus, and computer program product for classifying user search data |
| CN108614832A (en) * | 2016-12-12 | 2018-10-02 | 苏宁云商集团股份有限公司 | A kind of user individual commercial articles searching implementation method and device |
| CN110706055A (en) * | 2019-08-22 | 2020-01-17 | 平安科技（深圳）有限公司 | Commodity information pushing method and device, storage medium and computer equipment |
| WO2020018812A1 (en) * | 2018-07-18 | 2020-01-23 | The Dun & Bradstreet Corporation | Artificial intelligence engine for generating semantic directions for websites for automated entity targeting to mapped identities |
| CN110889747A (en) * | 2019-12-02 | 2020-03-17 | 腾讯科技（深圳）有限公司 | Commodity recommendation method, commodity recommendation device, commodity recommendation system, computer equipment and storage medium |
| CN111078842A (en) * | 2019-12-31 | 2020-04-28 | 北京每日优鲜电子商务有限公司 | Method, device, server and storage medium for determining query result |
| CN111241844A (en) * | 2018-11-29 | 2020-06-05 | 北京搜狗科技发展有限公司 | Information recommendation method and device |
| CN111260428A (en) * | 2018-12-03 | 2020-06-09 | 北京京东尚科信息技术有限公司 | Commodity recommendation method and device |
-
2020
- 2020-07-20 CN CN202010699655.8A patent/CN111737418B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20180268438A1 (en) * | 2012-10-04 | 2018-09-20 | Groupon, Inc. | Method, apparatus, and computer program product for classifying user search data |
| CN108614832A (en) * | 2016-12-12 | 2018-10-02 | 苏宁云商集团股份有限公司 | A kind of user individual commercial articles searching implementation method and device |
| WO2020018812A1 (en) * | 2018-07-18 | 2020-01-23 | The Dun & Bradstreet Corporation | Artificial intelligence engine for generating semantic directions for websites for automated entity targeting to mapped identities |
| CN111241844A (en) * | 2018-11-29 | 2020-06-05 | 北京搜狗科技发展有限公司 | Information recommendation method and device |
| CN111260428A (en) * | 2018-12-03 | 2020-06-09 | 北京京东尚科信息技术有限公司 | Commodity recommendation method and device |
| CN110706055A (en) * | 2019-08-22 | 2020-01-17 | 平安科技（深圳）有限公司 | Commodity information pushing method and device, storage medium and computer equipment |
| CN110889747A (en) * | 2019-12-02 | 2020-03-17 | 腾讯科技（深圳）有限公司 | Commodity recommendation method, commodity recommendation device, commodity recommendation system, computer equipment and storage medium |
| CN111078842A (en) * | 2019-12-31 | 2020-04-28 | 北京每日优鲜电子商务有限公司 | Method, device, server and storage medium for determining query result |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112328889A (en) * | 2020-11-23 | 2021-02-05 | 北京字节跳动网络技术有限公司 | Method and device for determining recommended search terms, readable medium and electronic equipment |
| KR20210151017A (en) * | 2020-11-24 | 2021-12-13 | 베이징 바이두 넷컴 사이언스 테크놀로지 컴퍼니 리미티드 | Method and apparatus for training search model, and method and apparatus for searching for target object |
| JP2022024102A (en) * | 2020-11-24 | 2022-02-08 | ベイジン バイドゥ ネットコム サイエンス テクノロジー カンパニー リミテッド | Method for training search model, method for searching target object and device therefor |
| KR102685008B1 (en) * | 2020-11-24 | 2024-07-12 | 베이징 바이두 넷컴 사이언스 테크놀로지 컴퍼니 리미티드 | Method and apparatus for training search model, and method and apparatus for searching for target object |
| CN113111800A (en) * | 2021-04-19 | 2021-07-13 | 北京每日优鲜电子商务有限公司 | Commodity shelf-loading judging method and device for unmanned container and storage medium |
| JP2022173084A (en) * | 2021-05-06 | 2022-11-17 | ネイバー コーポレーション | Method for searching for item based on burying similarity, computer device, and computer program |
| JP7343649B2 (en) | 2021-05-06 | 2023-09-12 | ネイバー コーポレーション | Product search method, computer device, and computer program based on embedded similarity |
| CN114282965A (en) * | 2021-12-21 | 2022-04-05 | 唯品会(广州)软件有限公司 | Recall scheme evaluation method and device and computer equipment |
| CN114139041A (en) * | 2022-01-28 | 2022-03-04 | 浙江口碑网络技术有限公司 | Category relevance prediction network training and category relevance prediction method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111737418B (en) | 2021-05-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111737418B (en) | Method, apparatus and storage medium for predicting relevance of search term and commodity | |
| CN108629665B (en) | Personalized commodity recommendation method and system | |
| US20210027146A1 (en) | Method and apparatus for determining interest of user for information item | |
| CN111784455B (en) | Article recommendation method and recommendation equipment | |
| US11995112B2 (en) | System and method for information recommendation | |
| CN104866474B (en) | Individuation data searching method and device | |
| CN111461841B (en) | Article recommendation method, device, server and storage medium | |
| CN102542012B (en) | The result of classified search inquiry | |
| CN102902691B (en) | Recommend method and system | |
| TWI615724B (en) | Information push, search method and device based on electronic information-based keyword extraction | |
| CN110196904B (en) | Method and device for acquiring recommendation information and computer readable storage medium | |
| CN111738805B (en) | Behavior log-based search recommendation model generation method, device and storage medium | |
| TW201520790A (en) | Individualized data search | |
| CN106557480B (en) | Method and device for realizing query rewriting | |
| CN108205768A (en) | Database building method and data recommendation method and device, equipment and storage medium | |
| US11682060B2 (en) | Methods and apparatuses for providing search results using embedding-based retrieval | |
| CN111400613A (en) | Article recommendation method, device, medium and computer equipment | |
| CN111967924A (en) | Commodity recommendation method, commodity recommendation device, computer device, and medium | |
| CN110264277A (en) | Data processing method and device, medium and the calculating equipment executed by calculating equipment | |
| CN110795613A (en) | Commodity searching method, device and system and electronic equipment | |
| CN118250516B (en) | Hierarchical processing method for users | |
| CN117648481A (en) | Sample processing method and device of recommendation model, electronic equipment and storage medium | |
| CN114022233A (en) | Novel commodity recommendation method | |
| CN111460300B (en) | Network content pushing method, device and storage medium | |
| CN112989020A (en) | Information processing method, apparatus and computer readable storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20211126 Address after: Room 084, No. 1-309, 3rd floor, commercial building, No. 9 Wangjing street, Chaoyang District, Beijing 100102 Patentee after: Beijing Daily Youxian Technology Co.,Ltd. Address before: 100102 room 801, 08 / F, building 7, yard 34, Chuangyuan Road, Chaoyang District, Beijing Patentee before: BEIJING MISSFRESH E-COMMERCE Co.,Ltd. |
|
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20231109 Address after: C1-331 Zheshang Pedestrian Street, Urban Rural Integration Demonstration Zone, Hebi City, Henan Province, 458000 Patentee after: Hebi Shiyu Qiluo E-commerce Co.,Ltd. Address before: Room 084, No. 1-309, 3rd floor, commercial building, No. 9 Wangjing street, Chaoyang District, Beijing 100102 Patentee before: Beijing Daily Youxian Technology Co.,Ltd. |
|
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20231129 Address after: 310000, Room 4251, Building 7, No. 1197 Bin'an Road, Changhe Street, Binjiang District, Hangzhou City, Zhejiang Province, China (Zhejiang) Pilot Free Trade Zone Patentee after: Hangzhou Zhongzhou Zhilian Technology Co.,Ltd. Address before: C1-331 Zheshang Pedestrian Street, Urban Rural Integration Demonstration Zone, Hebi City, Henan Province, 458000 Patentee before: Hebi Shiyu Qiluo E-commerce Co.,Ltd. |
|
| TR01 | Transfer of patent right |