CN112488781A - Search recommendation method and device, electronic equipment and readable storage medium - Google Patents
Search recommendation method and device, electronic equipment and readable storage medium Download PDFInfo
- Publication number
- CN112488781A CN112488781A CN202011249569.3A CN202011249569A CN112488781A CN 112488781 A CN112488781 A CN 112488781A CN 202011249569 A CN202011249569 A CN 202011249569A CN 112488781 A CN112488781 A CN 112488781A
- Authority
- CN
- China
- Prior art keywords
- information
- candidate
- candidate information
- store
- query
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 64
- 238000012163 sequencing technique Methods 0.000 claims abstract description 24
- 238000006243 chemical reaction Methods 0.000 claims description 30
- 230000006399 behavior Effects 0.000 claims description 28
- 238000012216 screening Methods 0.000 claims description 14
- 238000004364 calculation method Methods 0.000 claims description 9
- 230000015654 memory Effects 0.000 claims description 6
- 238000000605 extraction Methods 0.000 claims description 4
- 238000004590 computer program Methods 0.000 claims description 3
- 230000008569 process Effects 0.000 abstract description 16
- 241000238557 Decapoda Species 0.000 description 18
- 241000733943 Hapalogaster mertensii Species 0.000 description 17
- 235000013336 milk Nutrition 0.000 description 7
- 239000008267 milk Substances 0.000 description 7
- 210000004080 milk Anatomy 0.000 description 7
- 238000013528 artificial neural network Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 6
- 238000010801 machine learning Methods 0.000 description 5
- 238000012549 training Methods 0.000 description 5
- 235000008429 bread Nutrition 0.000 description 4
- 230000001960 triggered effect Effects 0.000 description 4
- 238000005457 optimization Methods 0.000 description 3
- 235000013311 vegetables Nutrition 0.000 description 3
- 241000251468 Actinopterygii Species 0.000 description 2
- 241000157468 Reinhardtius hippoglossoides Species 0.000 description 2
- 230000009471 action Effects 0.000 description 2
- 238000013475 authorization Methods 0.000 description 2
- 238000004422 calculation algorithm Methods 0.000 description 2
- 238000013527 convolutional neural network Methods 0.000 description 2
- 239000000645 desinfectant Substances 0.000 description 2
- 235000013305 food Nutrition 0.000 description 2
- 239000013505 freshwater Substances 0.000 description 2
- 238000007477 logistic regression Methods 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- 230000011218 segmentation Effects 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 241000238017 Astacoidea Species 0.000 description 1
- 241000058338 Macrobrachium nipponense Species 0.000 description 1
- 241000404975 Synchiropus splendidus Species 0.000 description 1
- 235000020247 cow milk Nutrition 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 229910052900 illite Inorganic materials 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003062 neural network model Methods 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
- 230000008520 organization Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000000306 recurrent effect Effects 0.000 description 1
- 230000006403 short-term memory Effects 0.000 description 1
- 235000013618 yogurt Nutrition 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
- G06F16/90—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
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Physics & Mathematics (AREA)
- Economics (AREA)
- General Business, Economics & Management (AREA)
- Strategic Management (AREA)
- Marketing (AREA)
- Development Economics (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The embodiment of the disclosure provides a search recommendation method, a search recommendation device, an electronic device and a readable storage medium, wherein the method comprises the following steps: responding to a query request, and acquiring query information carried in the query request, and user information and store information corresponding to the query information; recalling a set of candidate information based on the query information, the user information, and the store information; and sequencing each candidate information in the candidate information set to obtain a return result corresponding to the query request. The embodiment of the disclosure can improve the searching efficiency, simplify the operation process of user searching and improve the user experience.
Description
Technical Field
The embodiment of the disclosure relates to the technical field of internet, and in particular relates to a search recommendation method and device, an electronic device and a readable storage medium.
Background
With the development of electronic commerce platforms, the forms of displaying, publicizing and selling commodities through the internet tend to be more and more normalized. The merchant can input commodity information in the e-commerce platform, so that the commodity information can be displayed to the user through the e-commerce platform, and the user can search commodities provided by the merchant on line.
At present, search results are mainly pushed to a user according to keywords in query information, and the pushed search results are all search results containing the keywords, for example, for the keyword "live mitten crab", the pushed search results are all search results containing the keyword "live mitten crab", and are directly related in a word.
However, when a user inputs query information in a store of a certain merchant to search for a commodity, the conditions of no search result and few search results often occur due to the inventory and supply problems of the commodity in the store, and if the user finds that the current search result cannot meet the requirement, the user needs to return to the previous-level page to readjust the query information to search. The user needs to continuously try and make mistakes in the repeated process of browsing the search results and adjusting the query information to find the search results meeting the self requirements, so that the search efficiency is low, the operation process is complicated, and the user experience is influenced.
Disclosure of Invention
Embodiments of the present disclosure provide a search recommendation method, a search recommendation device, an electronic device, and a readable storage medium, so as to improve search efficiency, simplify an operation process of user search, and improve user experience.
According to a first aspect of embodiments of the present disclosure, there is provided a search recommendation method, the method including:
responding to a query request, and acquiring query information carried in the query request, and user information and store information corresponding to the query information;
recalling a set of candidate information based on the query information, the user information, and the store information;
and sequencing each candidate information in the candidate information set to obtain a return result corresponding to the query request.
According to a second aspect of embodiments of the present disclosure, there is provided a search recommendation apparatus including:
the information acquisition module is used for responding to an inquiry request and acquiring inquiry information carried in the inquiry request, and user information and store information corresponding to the inquiry information;
a candidate recall module for recalling a set of candidate information based on the query information, the user information, and the store information;
and the result determining module is used for sequencing all candidate information in the candidate information set to obtain a return result corresponding to the query request.
According to a third aspect of embodiments of the present disclosure, there is provided an electronic apparatus including:
a processor, a memory, and a computer program stored on the memory and executable on the processor, wherein the processor implements the aforementioned search recommendation method when executing the program.
According to a fourth aspect of embodiments of the present disclosure, there is provided a readable storage medium, wherein instructions, when executed by a processor of an electronic device, enable the electronic device to perform the aforementioned search recommendation method.
The embodiment of the disclosure provides a search recommendation method, a search recommendation device, an electronic device and a readable storage medium, wherein the method comprises the following steps:
the method comprises the steps of responding to a query request, and acquiring query information carried in the query request, and user information and store information corresponding to the query information; recalling a set of candidate information based on the query information, the user information, and the store information; and sequencing each candidate information in the candidate information set to obtain a return result corresponding to the query request.
According to the user information, the information such as the historical behaviors and the interest habits of the user can be acquired, and then the commodities which accord with the preference of the user can be recalled to serve as candidate information. Conversion information such as the sales volume of the store can be acquired from the store information, and a product with a high sales volume in the store can be recalled as candidate information. Thus, the embodiment of the disclosure, based on the query information, the user information, and the store information, the candidate information included in the recalled candidate information set includes not only the candidate information related to the query information, but also the candidate information related to the user information (such as the goods according with the user preference), and candidate information related to store information (such as products with high sales in stores) can reflect the search tendency intention of the user, and therefore, in the case of less recall results of the query information, the embodiment of the disclosure can actively provide the user with the recommendation information more meeting the search requirements of the user, the method and the device can avoid continuous trial and error of the user in the repeated process of browsing the search result and adjusting the query information, improve the search efficiency, simplify the operation process of user search and improve the user experience.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the drawings needed to be used in the description of the embodiments of the present disclosure will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present disclosure, and it is obvious for those skilled in the art that other drawings can be obtained according to the drawings without inventive exercise.
FIG. 1 shows a flow diagram of the steps of a search recommendation method in one embodiment of the present disclosure;
FIG. 2 shows a schematic flow diagram of an example of obtaining a set of candidate information in one embodiment of the present disclosure;
FIG. 3 shows a block diagram of a search recommendation device in one embodiment of the present disclosure;
fig. 4 shows a block diagram of an electronic device provided by an embodiment of the present disclosure.
Detailed Description
Technical solutions in the embodiments of the present disclosure will be clearly and completely described below with reference to the drawings in the embodiments of the present disclosure, and it is apparent that the described embodiments are some, but not all, of the embodiments of the present disclosure. All other embodiments, which can be obtained by a person skilled in the art without making creative efforts based on the embodiments of the present disclosure, belong to the protection scope of the embodiments of the present disclosure.
Example one
Referring to fig. 1, a flow diagram of the steps of a search recommendation method in one embodiment of the present disclosure is shown, the method comprising:
and 103, sequencing all candidate information in the candidate information set to obtain a return result corresponding to the query request.
The search recommendation method provided by the present disclosure is applicable to electronic devices including, but not limited to: smart phones, tablet computers, electronic book readers, MP3 (Moving Picture Experts Group Audio Layer III) players, MP4 (Moving Picture Experts Group Audio Layer IV) players, laptop portable computers, car-mounted computers, desktop computers, set-top boxes, smart televisions, wearable devices, and the like.
The electronic device may be installed with a client, and the client may be an APP (Application, APP for short) or a web browser used by the world wide web, and the like. The client can be used for displaying commodity search interfaces of different merchants, a user inputs query information (query) in a search box of the commodity search interface of a certain merchant and can trigger a query request by clicking a search button, and a search engine searches search results matched with the query information in commodities provided by the merchant according to a certain search strategy and returns the search results to the user.
The embodiment of the disclosure can be used for solving the problem of no result or few results in a search scene. In order to avoid the situation of no result and few result in the search, after receiving the query request, the embodiment of the present disclosure responds to the query request, and first obtains the query information carried in the query request, and the user information and store information corresponding to the query information.
It should be noted that, the specific type of the query information is not limited by the embodiments of the present disclosure. The query information may be text information or voice information. When the query information is voice information, voice recognition can be performed on the voice information to obtain corresponding text information. The user information may include: user identification, username, user fingerprint, etc. may be used to uniquely identify the user. The store information may include: store identification, store name, etc. may be used to uniquely identify the store. Further, the store information may further include: store information of the associated store corresponding to the store identification within the preset range. The associated store refers to other stores having an association relation with the store corresponding to the store identification, such as a chain store. The preset range can be set according to actual requirements, for example, the preset range can be a designated area, city, country, and the like.
In one example, user A enters the query information "hairy crab" in the search interface of store X of merchant A and triggers the query request. The embodiment of the disclosure may respond to the query request, and obtain query information (e.g., "hairy crab") carried in the query request, and user information (e.g., user identifier of user a) and store information (e.g., store identifier of store X) corresponding to the query information. Further, the embodiment of the present disclosure may further obtain store information of a store associated with the store X in the preset range of the merchant a, such as store information of chain stores (e.g., store Y and store Z) of the store X in the sunny district.
According to the user information, the information such as the historical behaviors and the interest habits of the user can be acquired, and then the commodities which accord with the preference of the user can be recalled to serve as candidate information. Conversion information such as the sales volume of the store can be acquired from the store information, and a product with a high sales volume in the store can be recalled as candidate information. Thus, the embodiment of the present disclosure, based on the query information, the user information, and the store information, the candidate information included in the recalled candidate information set may include not only candidate information related to the query information, but also candidate information related to the user information (for example, a product that meets the user preference, etc.), and candidate information related to the store information (for example, a product with a higher sales volume in the store, etc.).
And finally, sequencing all candidate information in the candidate information set to obtain a return result corresponding to the query request. The returned results comprise search results recalled based on multiple query dimensions (query information dimensions, user information dimensions and store information dimensions), so that the condition that no results or few results are recalled in the query information can be avoided, and the accuracy and recall rate of search recommendation are improved.
In an optional embodiment of the present disclosure, the recalling a set of candidate information based on the query information, the user information, and the store information at step 102 comprises:
step S11, recalling a first candidate information set related to the query information based on the query information;
step S12, recalling a second candidate information set related to the user information based on the user information;
step S13, recalling a third candidate information set related to the store information based on the store information;
step S14, merging the first candidate information set, the second candidate information set, and the third candidate information set to obtain a candidate information set.
The disclosed embodiments recall a set of candidate information based on a plurality of query dimensions after receiving a query request. The plurality of query dimensions include a query information dimension, a user information dimension, and a store information dimension. The candidate information set comprises candidate information recalled by a plurality of query dimensions.
Specifically, based on the query information, a first set of candidate information related to the query information is recalled. The candidate information in the first candidate information set has strong correlation with the query information, and the correlation may include literal correlation, semantic correlation, category correlation, and the like. For example, after receiving the query request, in response to the query request, it is acquired that the query information is "live crab", the user information is "user a", and the store information is "store X". For the query information "hairy crab", the recalled first candidate information set may include the commodity "hairy crab", and may also include commodities related to the category of the commodity "hairy crab", such as "river crab".
Recalling a second candidate information set related to the user information based on the user information. The candidate information in the second candidate information set has a strong correlation with the user information, and the correlation may include historical behavior correlation, interest preference correlation, and the like. For example, for the user information "user a", the recalled second candidate information may include items that user a has recently purchased, and may also include items that user a prefers, such as items that user a frequently likes to purchase, and the like.
Recalling a third candidate information set related to the store information based on the store information. The candidate information in the third candidate information set has strong correlation with the store information, and the correlation may include sales volume correlation, click rate correlation, order placement rate correlation, promotion correlation and the like. For example, for the store information "store X", the third candidate information set recalled may include the item sold the highest in the store X, the item with the greatest promotion strength in the store X, and the like.
After the first candidate information set, the second candidate information set and the third candidate information set are recalled, the first candidate information set, the second candidate information set and the third candidate information set are merged and deduplicated, so that a candidate information set corresponding to the query request can be obtained.
In an optional embodiment of the present disclosure, the recalling, based on the query information, the first candidate information set related to the query information in step S11 includes:
s111, extracting a first keyword in the query information;
step S112, retrieving the first keyword and at least one piece of relevant dimension information corresponding to the first keyword in the inventory of the store corresponding to the query request to obtain a first recall result;
and S113, sorting the first recall result according to a preset screening condition to obtain a first candidate information set.
The first keyword may be a real word having a specific meaning, including a name of a person, a place name, a name of an organization, a proper noun, and the like. It should be noted that, in the embodiments of the present disclosure, the first and second terms are only used for distinguishing different nouns, and do not indicate the importance and the order of the nouns.
In one example, a query request is received for search query information "live crabs" triggered by user a at a search interface of store X. The query information comprises a first keyword 'hairy crab', and the first keyword and at least one piece of relevant dimension information corresponding to the first keyword are retrieved in the inventory of a store (store X) corresponding to the query request to obtain a first recall result. At least one piece of relevant dimension information corresponding to the first keyword may include a category, a brand, a label, and the like corresponding to the first keyword. At least one piece of relevant dimension information corresponding to the first keyword can be determined according to historical behavior data of a large number of users.
The historical behavior data of a large number of users can be obtained according to the purchase logs of a large number of users, for example, the purchase logs of a large number of users can be obtained, and the categories of commodities which are finally ordered after most of the users search the query word 'hairy crabs' are analyzed. For example, most users search for the query term "live crabs", and the final items to be placed include "live crabs", "river crabs", and the like, and the category to which the placed items belong is "crab category". Thus, when the search query information input by a single user is 'hairy crab', the category preference of 'hairy crab' can be determined to be 'crab category'. Thus, the first recall result may include merchandise related to "crabs", such as "river crabs", "bread crabs", and the like. For another example, it may be determined that the brand preference of most users is "Yangcheng lake" when searching for the query information "hairy crab" according to historical behavior data of a large number of users, and the first recall result may further include a commodity related to the brand preference of "hairy crab", such as "Yangcheng lake hairy crab".
And then, sorting the first recall result according to a preset screening condition to obtain a first candidate information set. It is understood that the preset screening condition is not limited by the embodiments of the present disclosure, for example, the preset screening condition may be a sales volume, a placing rate, a click rate, and the like. Optionally, the number of the candidate information included in the first candidate information set is a first number, and the first number is obtained by presetting, and the first number may be artificially set according to an actual demand or obtained by learning according to a machine learning model.
Further, in practical applications, there may be a case where the query information is long, and the query information may include a plurality of keywords. Therefore, the embodiment of the disclosure performs word segmentation on the query information, and extracts at least one first keyword.
For example, for the query information "cow milk", the first keyword "cow" and the first keyword "milk" may be extracted after word segmentation. The query information embodies brand preference (Mongolian cow) and category preference (pure milk), and preference characteristics of each first keyword can be considered during retrieval. According to the first keyword ' Mongolian cow ' and the first keyword ' pure milk ', the commodity ' Mongolian cow pure milk ' can be recalled, other commodities of the brand of the Mongolian cow ' such as ' Mongolian cow yoghourt ' and the like, and pure milk of other brands such as ' illite pure milk ' and the like can also be recalled. In this way, in the case that the store 'Mongolian pure milk' is in short stock, other commodities which accord with the brand preference and the category preference of the user can be recommended to the user.
In an optional embodiment of the present disclosure, the user information may include a user identifier, and the recalling a second candidate information set related to the user information based on the user information in step S12 includes:
step S121, obtaining historical behavior information of the user identification in a preset time, wherein the historical behavior information comprises one or more of the following items: historical search information, historical click information and historical purchase information;
step S122, extracting a second keyword in the historical behavior information;
step S123, retrieving the second keyword and at least one piece of relevant dimension information corresponding to the second keyword in the inventory of the store corresponding to the query request to obtain a second recall result;
and step S124, sorting the second recall result according to preset screening conditions to obtain a second candidate information set.
According to the method and the device, the historical behavior information of the user in the preset time can be acquired according to the user identification, and it should be noted that before the user identification is acquired and the historical behavior information of the user identification in the preset time is acquired, the authorization of the user can be acquired firstly, and after the authorization of the user, the user identification and the historical behavior information of the user identification in the preset time can be acquired, so that the privacy safety of user data is ensured.
The preset time may be set according to needs, for example, the preset time may be a period of time indicating the near future, such as within the past week, within the past three days, and the like. In one example, historical search information, historical click information, historical purchase information, etc. of the user over the past week may be obtained. Still take the above-mentioned query request of receiving the search query information "live crabs" triggered by the user a at the search interface of the store X. The historical behavior information of the user A can be obtained according to the user identification of the user A, and if the historical behavior information of the user A is analyzed, the fact that the user A purchases the weever in the past week can be known, and the second keyword can be extracted to be the weever.
And in the inventory of the store (store X) corresponding to the query request, retrieving the second keyword (weever) and at least one piece of relevant dimension information corresponding to the second keyword to obtain a second recall result. At least one piece of relevant dimension information corresponding to the second keyword may include a category, a brand, a label, and the like corresponding to the second keyword. At least one piece of relevant dimension information corresponding to the second keyword can be determined according to historical behavior information of the user. For example, if it is known from the historical behavior information of the user a that the user a purchased "weever" in the past week, the second recall result may include the product "weever". In addition, since the category to which the "weever" belongs is "fresh water fish", the second recall result may include other commodities of "fresh water fish" such as "mandarin fish", "turbot", and the like, in addition to the commodity "weever".
And sorting the second recall results according to preset screening conditions to obtain a second candidate information set. The preset screening conditions are not limited in the embodiment of the disclosure, and the preset screening conditions may be sales volume, click rate, and the like. Optionally, the number of the candidate information included in the second candidate information set is a second number, the second number is obtained by presetting, and the second number may be artificially set according to actual needs or obtained by learning according to a machine learning model.
According to the embodiment of the invention, the inventory characteristics of stores are considered in the search recommendation process, and the problem that partial query information has long tails, so that the search has no result or few results can be solved. For example, the query information is "clew", and the user behavior under the query information is sparse. The retrieval is carried out based on the 'vegetable group' and at least one piece of relevant dimension information corresponding to the 'vegetable group', the relevant dimension is taken as the category, and the category to which the query information 'vegetable group' belongs is 'refrigerated food', so that other recommended commodities of 'refrigerated food' can be included in the second recall result, and the condition that no result or few results are searched can be avoided.
In an optional embodiment of the present disclosure, the store information may include a store identifier, and the step S13 of recalling a third candidate information set related to the store information based on the store information includes:
step S131, conversion information of each commodity corresponding to the store identification is obtained, wherein the conversion information comprises one or more of the following items: click rate, order placement rate, and sales volume;
and S132, sequencing the commodities corresponding to the store identification according to the conversion information to obtain a third candidate information set.
Still take the above-mentioned query request of receiving the search query information "live crabs" triggered by the user a at the search interface of the store X. Acquiring conversion information of each commodity corresponding to a store identifier of a store X, wherein the conversion information comprises one or more of the following items: click rate, order placement rate, sales volume, etc. For convenience of description, taking the order placing rate of each commodity as an example, the commodities of the merchant B are sorted according to the order placing rate, and n (n is a positive integer) commodities with the highest order placing rate are added into the third candidate information set. Optionally, the number of the candidate information included in the third candidate information set is a third number, and the third number is obtained by presetting, and the third number may be artificially set according to actual needs, or obtained by learning according to a machine learning model. Wherein the value of n is less than or equal to the third number.
Optionally, the obtaining of the conversion information of each product corresponding to the store identifier in step S131 may include: and acquiring conversion information of the associated store corresponding to the store identification within a preset range. The associated store refers to other stores having an association relation with the store corresponding to the store identification, such as a chain store. In one example, user A searches for commodity X1 at store X, but since X1 is a new commodity at store X, the user data is sparse, and it is difficult to obtain conversion information of X1 at store X at the time of recommendation. In this case, the present disclosure may acquire conversion information of a product X1 in a store associated with store X within a preset range, and for example, acquire conversion information of a product X1 in store Y, the conversion information of a product X1 in store Y may be used as conversion information of a product X1 in store X.
It should be noted that the preset range may be set according to actual requirements, for example, the preset range may be a designated area, a city, a country, and the like.
Referring to fig. 2, a schematic flow chart illustrating an example of the above-described acquisition of the candidate information set in the embodiment of the present disclosure is shown. As shown in fig. 2, the user a searches for the query information "live crabs" at the store X as an example. After the first candidate information set, the second candidate information set and the third candidate information are obtained, the first candidate information set, the second candidate information set and the third candidate information set are merged to obtain a candidate information set. And finally, sequencing all candidate information in the candidate information set to obtain a return result corresponding to the query request.
In an optional embodiment of the present disclosure, the sorting the candidate information in the candidate information set in step 103 to obtain a returned result corresponding to the query request includes:
step S31, acquiring at least one reference feature corresponding to each candidate information in the candidate information set;
step S32, according to the at least one reference feature, ranking each candidate information in the candidate information set to obtain a return result corresponding to the query request.
In order to enable the returned result displayed to the user to meet the user requirement, the embodiment of the present disclosure ranks the candidate information in the candidate information set based on at least one reference feature, and returns top (k is a positive integer) ranked candidate information as the returned result to the user.
The at least one reference feature may comprise one or more of: query preference features, user preference features, merchandise preference features, search result preference features, and the like.
Wherein, the query preference feature may include: the number of clicks of a certain commodity clicked by the user under certain query information, or the number of orders of a certain commodity placed by the user under certain query information, and the like. The user preference features may include: the total number of orders placed by users of a certain commodity, the total number of clicks made by users of a certain commodity and the like. The merchandise preference features may include: inventory of goods in stores, conversion rates, etc. The search result preference feature may include: conversion rate of each search result under certain query information.
In the process of acquiring the candidate information, the relevant characteristics of the search information, the user information and the store information are fully considered. Specifically, by taking the above example as an example, the embodiments of the present disclosure may determine the category characteristics of the query information "live crabs" according to the user information. In the case that the query information is 'hairy crab', most users purchase 'crab class' commodities, so that the 'hairy crab' is more preferable to the 'crab class' in terms of the obtained 'hairy crab'. Thus, embodiments of the present disclosure may give a higher query preference feature score for "crab-like" goods under the query information "live crabs". Similarly, the related characteristics of the preference of the user A and the related characteristics of the preference of the store X goods can be obtained.
The relevant characteristics of the user A preference can include brand preference, category preference and the like of the user A, and the relevant characteristics of the store X commodity preference can include sales volume, order placing rate, click rate and the like of commodities in the store X.
In addition, because the search recommendation and the search result have close relation, the search result preference characteristics can be considered in the process of making the search recommendation. For example, when the candidate commodities obtained under the query information "live action crabs" are ranked, the characteristics such as the conversion rate of the search results under the query information "live action crabs" can be considered.
In the search recommendation process, based on a plurality of query dimensions of the query information, the user information and the store information, the finally obtained return result not only comprises the commodity information closely related to the query information, but also comprises the recommended commodity information highly related to the user information and the store information. Therefore, under the condition that the number of the query information recall results is small, the substitute meeting the requirements of the user can be recommended to the user. Under the condition that the number of the query information recalls is large, the embodiment of the disclosure can recommend more accurate search results to the user. Further, the search results obtained based on the plurality of query dimensions may further include a collocation recommended to the user. For example, when the query information is "disposable mask", it can be known that most users have behavior of purchasing "disinfectant" while purchasing "disposable mask" based on the user information dimension, and the finally obtained return result includes not only the product information closely related to the query information but also the recommended product "disinfectant".
Therefore, the searching and recommending method and device based on the multi-dimension are used for searching and recommending, the searching relevance can be guaranteed, the user experience is improved, and the conversion rate can be improved by considering the user preference and the popularity information of the commodities.
In an optional embodiment of the present disclosure, the step S32 of sorting, according to the at least one reference feature, each candidate information in the candidate information set to obtain a returned result corresponding to the query request includes:
step S321, calculating a feature score corresponding to each reference feature in the at least one reference feature for each candidate information in the candidate information set;
step S322, performing weighted calculation on the feature score of each reference feature corresponding to each candidate information to obtain a reference score corresponding to each candidate information;
step S323, sorting each candidate information in the candidate information set according to the reference score to obtain a sorting result;
step S324, determining candidate information meeting a preset sorting condition in the sorting result as a returned result corresponding to the query request.
Still take the above-mentioned query request of receiving the search query information "live crabs" triggered by the user a at the search interface of the store X. The candidate information set corresponding to the query request is assumed to include the following candidate information: river crab, bread crab, crayfish, weever, turbot, etc.
In this example, the query preference feature, the user preference feature, and the commodity preference feature are selected as reference features used when ranking the candidate information. For each candidate information, the feature score corresponding to each reference feature can be obtained through calculation, and the reference score of each candidate information can be obtained through weighted calculation according to the feature score corresponding to each reference feature of each candidate information.
For example, for the candidate information "river crab", the candidate information has a feature score of s1 corresponding to the query preference feature, s2 corresponding to the user preference feature, and s3 corresponding to the commodity preference feature. And performing weighted calculation on s1, s2 and s3 to obtain the reference score of the candidate information 'river crab'. It will be appreciated that the weight corresponding to each reference feature may be set as desired.
Alternatively, the feature score of the candidate information corresponding to each reference feature may be determined according to the historical conversion features of the candidate information. The historical conversion characteristics may include: user down singular, user number of clicks, sales, etc. In one example, in the case where the user search query information "live crabs" is 10 units in the singular number, 2 units in the singular number, and 0 unit in the singular number, the user of "river crabs" is the user of "bread crabs". For the candidate information "river crab", s1 can be noted as 10; for the candidate information "bread crab", s1 can be noted as 2; for the candidate information "macrobrachium nipponensis", s1 may be noted as 0.
Referring to table 1, a feature score indication of a candidate message corresponding to each reference feature of the embodiment of the disclosure is shown.
TABLE 1
The first query preference feature can represent the user ordering number of the candidate information x under the condition that the query information is the hairy crab, the second query preference feature can represent the user click number of the candidate information x under the condition that the query information is the hairy crab, the first user preference feature can represent the user ordering total number of the candidate information x, and the second user preference feature can represent the user click total number of the candidate information x. The first commodity preference feature may represent the sales volume of the candidate information x in the current store, and the second commodity preference feature may represent the total sales volume of all associated stores of the candidate information x in the preset range. The candidate information x is the candidate information shown in the first column in table 1.
Of course, the above-mentioned manner of calculating the score of the candidate information corresponding to each reference feature is only an example of the disclosure, and the embodiment of the disclosure does not limit the specific manner of calculating the score of the candidate information corresponding to each reference feature. In addition, the embodiments of the present disclosure do not impose limitations on the number and types of reference features.
In the same method, a reference score can be calculated for each candidate information in the candidate information set, and each candidate information in the candidate information set is sorted according to the reference score to obtain a sorting result; and determining candidate information meeting preset sorting conditions in the sorting results as a return result corresponding to the query request. The predetermined sorting condition may be m before sorting (m is a positive integer).
In an optional embodiment of the present disclosure, the sorting, according to the at least one reference feature, each candidate information in the candidate information set includes: and inputting at least one reference characteristic corresponding to each candidate information in the candidate information set into a pre-trained ranking model, and outputting a ranking result through the ranking model.
In order to improve the accuracy and efficiency of ranking, the embodiments of the present disclosure may pre-train a ranking model, and input at least one reference feature corresponding to each candidate information in the candidate information set into the ranking model, that is, output a ranking result through the ranking model.
The ranking model may be obtained by supervised training of an existing neural network based on a large number of training samples and machine learning methods. It should be noted that, the embodiment of the present disclosure does not limit the model structure and the training method of the ranking model. The ranking model may fuse a variety of neural networks. The neural network includes, but is not limited to, at least one or a combination, superposition, nesting of at least two of the following: CNN (Convolutional Neural Network), LSTM (Long Short-Term Memory) Network, RNN (Simple Recurrent Neural Network), attention Neural Network, and the like.
Firstly, the embodiment of the disclosure can construct and initialize a sequencing model, and set model parameters of the initial model; and then, inputting the training samples into the initial model one by one, performing iterative optimization on the initial model according to the difference between the output result of the initial model and the labeled information in the training samples and a gradient descent algorithm, adjusting model parameters, stopping iterative optimization until the optimized model reaches a preset condition, and taking the model obtained by the last optimization as a trained sequencing model.
It should be noted that, in the embodiment of the present disclosure, the ranking model is taken as an example of a neural network model for explanation. In a specific implementation, the type of the ranking model is not limited, and the ranking model is not limited to a neural network, and may be all machine learning models, such as an LR (Logistic Regression) model, an XGBoost (electronic tree Gradient Boosting, or an integrated tree model).
In summary, in response to a query request, the embodiments of the present disclosure obtain query information carried in the query request, and user information and store information corresponding to the query information; recalling a set of candidate information based on the query information, the user information, and the store information; and sequencing each candidate information in the candidate information set to obtain a return result corresponding to the query request. According to the user information, the information such as the historical behaviors and the interest habits of the user can be acquired, and then the commodities which accord with the preference of the user can be recalled to serve as candidate information. Conversion information such as the sales volume of the store can be acquired from the store information, and a product with a high sales volume in the store can be recalled as candidate information. Thus, the embodiment of the disclosure, based on the query information, the user information, and the store information, the candidate information included in the recalled candidate information set includes not only the candidate information related to the query information, but also the candidate information related to the user information (such as the goods according with the user preference), and candidate information related to store information (such as products with high sales in stores) can reflect the search tendency intention of the user, and therefore, in the case of less recall results of the query information, the embodiment of the disclosure can actively provide the user with the recommendation information more meeting the search requirements of the user, the method and the device can avoid continuous trial and error of the user in the repeated process of browsing the search result and adjusting the query information, improve the search efficiency, simplify the operation process of user search and improve the user experience.
It is noted that, for simplicity of description, the method embodiments are described as a series of acts or combination of acts, but those skilled in the art will recognize that the disclosed embodiments are not limited by the described order of acts, as some steps may occur in other orders or concurrently with other steps in accordance with the disclosed embodiments. Further, those skilled in the art will appreciate that the embodiments described in the specification are presently preferred and that no particular act is required of the disclosed embodiments.
Example two
Referring to fig. 3, a block diagram of a search recommendation apparatus in one embodiment of the present disclosure is shown, specifically as follows.
The information acquisition module 301 is configured to respond to a query request, and acquire query information carried in the query request, and user information and store information corresponding to the query information;
a candidate recall module 302 configured to recall a set of candidate information based on the query information, the user information, and the store information;
and the result determining module 303 is configured to rank each candidate information in the candidate information set to obtain a return result corresponding to the query request.
Optionally, the candidate recall module 302 includes:
a first recall sub-module for recalling a first set of candidate information related to the query information based on the query information;
a second recall submodule configured to recall, based on the user information, a second candidate information set related to the user information;
a third recall sub-module for recalling a third candidate information set related to the store information based on the store information;
and the candidate merging submodule is used for merging the first candidate information set, the second candidate information set and the third candidate information set to obtain a candidate information set.
Optionally, the first recall sub-module includes:
the first extraction unit is used for extracting a first keyword in the query information;
the first recall unit is used for retrieving the first keyword and at least one piece of relevant dimension information corresponding to the first keyword in the inventory of the store corresponding to the query request to obtain a first recall result;
and the first sequencing unit is used for sequencing the first recall result according to a preset screening condition to obtain a first candidate information set.
Optionally, the user information includes a user identifier, and the second recall sub-module includes:
a first obtaining unit, configured to obtain historical behavior information of the user identifier within a preset time, where the historical behavior information includes one or more of the following items: historical search information, historical click information and historical purchase information;
the second extraction unit is used for extracting a second keyword in the historical behavior information;
the second recall unit is used for retrieving the second keyword and at least one piece of relevant dimension information corresponding to the second keyword in the inventory of the store corresponding to the query request to obtain a second recall result;
and the second sorting unit is used for sorting the second recall result according to a preset screening condition to obtain a second candidate information set.
Optionally, the store information includes a store identification, and the third recall sub-module includes:
a second obtaining unit, configured to obtain conversion information of each commodity corresponding to the store identifier, where the conversion information includes one or more of the following items: click rate, order placement rate, and sales volume;
and the third sorting unit is used for sorting the commodities corresponding to the store identification according to the conversion information to obtain a third candidate information set.
Optionally, the result determining module 303 includes:
the characteristic obtaining submodule is used for obtaining at least one reference characteristic corresponding to each candidate information in the candidate information set;
and the sequencing determining submodule is used for sequencing each candidate information in the candidate information set according to the at least one reference characteristic to obtain a return result corresponding to the query request.
Optionally, the rank determining sub-module includes:
a feature score calculation unit, configured to calculate, for each candidate information in the candidate information set, a feature score corresponding to each reference feature in the at least one reference feature respectively;
the reference score calculating unit is used for performing weighted calculation on the feature score of each reference feature corresponding to each candidate information to obtain a reference score corresponding to each candidate information;
the candidate sorting unit is used for sorting each candidate information in the candidate information set according to the reference score to obtain a sorting result;
and the result determining unit is used for determining that the candidate information meeting the preset sorting condition in the sorting result is a return result corresponding to the query request.
Optionally, the ranking determining sub-module is specifically configured to input at least one reference feature corresponding to each candidate information in the candidate information set into a pre-trained ranking model, and output a ranking result through the ranking model.
Based on the query information, the user information and the store information, the candidate information included in the recalled candidate information set includes not only candidate information related to the query information, but also candidate information related to the user information (such as commodities meeting user preferences) and candidate information related to the store information (such as commodities with higher sales in stores), which can better reflect the search tendency intention of the user.
For the device embodiment, since it is basically similar to the method embodiment, the description is simple, and for the relevant points, refer to the partial description of the method embodiment.
The embodiments in the present specification are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other.
With regard to the apparatus in the above-described embodiment, the specific manner in which each module performs the operation has been described in detail in the embodiment related to the method, and will not be elaborated here.
An embodiment of the present disclosure also provides an electronic device, referring to fig. 4, including: a processor 401, a memory 402 and a computer program 4021 stored on and executable on said memory, said processor implementing the search recommendation method of the previous embodiments when executing said program.
Embodiments of the present disclosure also provide a readable storage medium, in which instructions, when executed by a processor of an electronic device, enable the electronic device to perform the search recommendation method of the foregoing embodiments.
For the device embodiment, since it is basically similar to the method embodiment, the description is simple, and for the relevant points, refer to the partial description of the method embodiment.
The algorithms and displays presented herein are not inherently related to any particular computer, virtual machine, or other apparatus. Various general purpose systems may also be used with the teachings herein. The required structure for constructing such a system will be apparent from the description above. In addition, embodiments of the present disclosure are not directed to any particular programming language. It is appreciated that a variety of programming languages may be used to implement the teachings of the embodiments of the present disclosure as described herein, and any descriptions of specific languages are provided above to disclose the best modes of the embodiments of the present disclosure.
In the description provided herein, numerous specific details are set forth. However, it is understood that embodiments of the present disclosure may be practiced without these specific details. In some instances, well-known methods, structures and techniques have not been shown in detail in order not to obscure an understanding of this description.
Similarly, it should be appreciated that in the foregoing description of exemplary embodiments of the disclosure, various features of the embodiments of the disclosure are sometimes grouped together in a single embodiment, figure, or description thereof for the purpose of streamlining the disclosure and aiding in the understanding of one or more of the various inventive aspects. However, the disclosed method should not be interpreted as reflecting an intention that: that is, claimed embodiments of the disclosure require more features than are expressly recited in each claim. Rather, as the following claims reflect, inventive aspects lie in less than all features of a single foregoing disclosed embodiment. Thus, the claims following the detailed description are hereby expressly incorporated into this detailed description, with each claim standing on its own as a separate embodiment of an embodiment of this disclosure.
Those skilled in the art will appreciate that the modules in the device in an embodiment may be adaptively changed and disposed in one or more devices different from the embodiment. The modules or units or components of the embodiments may be combined into one module or unit or component, and furthermore they may be divided into a plurality of sub-modules or sub-units or sub-components. All of the features disclosed in this specification (including any accompanying claims, abstract and drawings), and all of the processes or elements of any method or apparatus so disclosed, may be combined in any combination, except combinations where at least some of such features and/or processes or elements are mutually exclusive. Each feature disclosed in this specification (including any accompanying claims, abstract and drawings) may be replaced by alternative features serving the same, equivalent or similar purpose, unless expressly stated otherwise.
The various component embodiments of the disclosure may be implemented in hardware, or in software modules running on one or more processors, or in a combination thereof. It will be appreciated by those skilled in the art that a microprocessor or Digital Signal Processor (DSP) may be used in practice to implement some or all of the functions of some or all of the components in a sequencing device according to embodiments of the present disclosure. Embodiments of the present disclosure may also be implemented as an apparatus or device program for performing a portion or all of the methods described herein. Such programs implementing embodiments of the present disclosure may be stored on a computer readable medium or may be in the form of one or more signals. Such a signal may be downloaded from an internet website or provided on a carrier signal or in any other form.
It should be noted that the above-mentioned embodiments illustrate rather than limit embodiments of the disclosure, and that those skilled in the art will be able to design alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. The word "comprising" does not exclude the presence of elements or steps not listed in a claim. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. Embodiments of the disclosure may be implemented by means of hardware comprising several distinct elements, and by means of a suitably programmed computer. In the unit claims enumerating several means, several of these means may be embodied by one and the same item of hardware. The usage of the words first, second and third, etcetera do not indicate any ordering. These words may be interpreted as names.
It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described systems, apparatuses and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
The above description is only for the purpose of illustrating the preferred embodiments of the present disclosure and is not to be construed as limiting the embodiments of the present disclosure, and any modifications, equivalents, improvements and the like that are made within the spirit and principle of the embodiments of the present disclosure are intended to be included within the scope of the embodiments of the present disclosure.
The above description is only a specific implementation of the embodiments of the present disclosure, but the scope of the embodiments of the present disclosure is not limited thereto, and any person skilled in the art can easily conceive of changes or substitutions within the technical scope of the embodiments of the present disclosure, and all the changes or substitutions should be covered by the scope of the embodiments of the present disclosure. Therefore, the protection scope of the embodiments of the present disclosure shall be subject to the protection scope of the claims.
Claims (18)
1. A search recommendation method, the method comprising:
responding to a query request, and acquiring query information carried in the query request, and user information and store information corresponding to the query information;
recalling a set of candidate information based on the query information, the user information, and the store information;
and sequencing each candidate information in the candidate information set to obtain a return result corresponding to the query request.
2. The method of claim 1, wherein recalling a set of candidate information based on the query information, the user information, and the store information comprises:
recalling a first candidate information set related to the query information based on the query information;
recalling a second candidate information set related to the user information based on the user information;
recalling a third candidate information set related to the store information based on the store information;
and merging the first candidate information set, the second candidate information set and the third candidate information set to obtain a candidate information set.
3. The method of claim 1, wherein recalling a first set of candidate information related to the query information based on the query information comprises:
extracting a first keyword in the query information;
in the inventory of the store corresponding to the query request, retrieving the first keyword and at least one piece of relevant dimension information corresponding to the first keyword to obtain a first recall result;
and sorting the first recall results according to preset screening conditions to obtain a first candidate information set.
4. The method of claim 2, wherein the user information comprises a user identification, and wherein recalling a second candidate set of information related to the user information based on the user information comprises:
obtaining historical behavior information of the user identification in a preset time, wherein the historical behavior information comprises one or more of the following items: historical search information, historical click information and historical purchase information;
extracting a second keyword in the historical behavior information;
in the inventory of the store corresponding to the query request, retrieving the second keyword and at least one piece of relevant dimension information corresponding to the second keyword to obtain a second recall result;
and sorting the second recall results according to preset screening conditions to obtain a second candidate information set.
5. The method of claim 2, wherein the store information comprises a store identification, and wherein recalling a third set of candidate information related to the store information based on the store information comprises:
acquiring conversion information of each commodity corresponding to the store identification, wherein the conversion information comprises one or more of the following items: click rate, order placement rate, and sales volume;
and sequencing the commodities corresponding to the store identification according to the conversion information to obtain a third candidate information set.
6. The method according to claim 1, wherein the sorting the candidate information in the candidate information set to obtain the returned result corresponding to the query request comprises:
acquiring at least one reference feature corresponding to each candidate information in the candidate information set;
and sequencing each candidate information in the candidate information set according to the at least one reference characteristic to obtain a return result corresponding to the query request.
7. The method according to claim 6, wherein the sorting the candidate information in the candidate information set according to the at least one reference feature to obtain a returned result corresponding to the query request comprises:
respectively calculating a feature score corresponding to each reference feature in the at least one reference feature for each candidate information in the candidate information set;
performing weighted calculation on the feature score of each reference feature corresponding to each candidate information to obtain a reference score corresponding to each candidate information;
sorting each candidate information in the candidate information set according to the reference score to obtain a sorting result;
and determining candidate information meeting preset sorting conditions in the sorting results as a return result corresponding to the query request.
8. The method of claim 6, wherein the ranking each candidate information in the set of candidate information according to the at least one reference feature comprises:
and inputting at least one reference characteristic corresponding to each candidate information in the candidate information set into a pre-trained ranking model, and outputting a ranking result through the ranking model.
9. A search recommendation apparatus, characterized in that the apparatus comprises:
the information acquisition module is used for responding to an inquiry request and acquiring inquiry information carried in the inquiry request, and user information and store information corresponding to the inquiry information;
a candidate recall module for recalling a set of candidate information based on the query information, the user information, and the store information;
and the result determining module is used for sequencing all candidate information in the candidate information set to obtain a return result corresponding to the query request.
10. The apparatus of claim 9, wherein the candidate recall module comprises:
a first recall sub-module for recalling a first set of candidate information related to the query information based on the query information;
a second recall submodule configured to recall, based on the user information, a second candidate information set related to the user information;
a third recall sub-module for recalling a third candidate information set related to the store information based on the store information;
and the candidate merging submodule is used for merging the first candidate information set, the second candidate information set and the third candidate information set to obtain a candidate information set.
11. The apparatus of claim 10, wherein the first recall submodule comprises:
the first extraction unit is used for extracting a first keyword in the query information;
the first recall unit is used for retrieving the first keyword and at least one piece of relevant dimension information corresponding to the first keyword in the inventory of the store corresponding to the query request to obtain a first recall result;
and the first sequencing unit is used for sequencing the first recall result according to a preset screening condition to obtain a first candidate information set.
12. The apparatus of claim 10, wherein the user information comprises a user identification, and wherein the second recall sub-module comprises:
a first obtaining unit, configured to obtain historical behavior information of the user identifier within a preset time, where the historical behavior information includes one or more of the following items: historical search information, historical click information and historical purchase information;
the second extraction unit is used for extracting a second keyword in the historical behavior information;
the second recall unit is used for retrieving the second keyword and at least one piece of relevant dimension information corresponding to the second keyword in the inventory of the store corresponding to the query request to obtain a second recall result;
and the second sorting unit is used for sorting the second recall result according to a preset screening condition to obtain a second candidate information set.
13. The apparatus of claim 10, wherein the store information comprises a store identification, and wherein the third recall sub-module comprises:
a second obtaining unit, configured to obtain conversion information of each commodity corresponding to the store identifier, where the conversion information includes one or more of the following items: click rate, order placement rate, and sales volume;
and the third sorting unit is used for sorting the commodities corresponding to the store identification according to the conversion information to obtain a third candidate information set.
14. The apparatus of claim 9, wherein the result determination module comprises:
the characteristic obtaining submodule is used for obtaining at least one reference characteristic corresponding to each candidate information in the candidate information set;
and the sequencing determining submodule is used for sequencing each candidate information in the candidate information set according to the at least one reference characteristic to obtain a return result corresponding to the query request.
15. The apparatus of claim 14, wherein the rank determination sub-module comprises:
a feature score calculation unit, configured to calculate, for each candidate information in the candidate information set, a feature score corresponding to each reference feature in the at least one reference feature respectively;
the reference score calculating unit is used for performing weighted calculation on the feature score of each reference feature corresponding to each candidate information to obtain a reference score corresponding to each candidate information;
the candidate sorting unit is used for sorting each candidate information in the candidate information set according to the reference score to obtain a sorting result;
and the result determining unit is used for determining that the candidate information meeting the preset sorting condition in the sorting result is a return result corresponding to the query request.
16. The apparatus according to claim 14, wherein the ranking determining sub-module is configured to input at least one reference feature corresponding to each candidate information in the candidate information set into a pre-trained ranking model, and output a ranking result through the ranking model.
17. An electronic device, comprising:
processor, memory and computer program stored on the memory and executable on the processor, characterized in that the processor implements the search recommendation method according to any one of claims 1-8 when executing the program.
18. A readable storage medium, wherein instructions in the storage medium, when executed by a processor of an electronic device, enable the electronic device to perform the search recommendation method of any of claims 1-8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011249569.3A CN112488781A (en) | 2020-11-10 | 2020-11-10 | Search recommendation method and device, electronic equipment and readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011249569.3A CN112488781A (en) | 2020-11-10 | 2020-11-10 | Search recommendation method and device, electronic equipment and readable storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112488781A true CN112488781A (en) | 2021-03-12 |
Family
ID=74929381
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011249569.3A Pending CN112488781A (en) | 2020-11-10 | 2020-11-10 | Search recommendation method and device, electronic equipment and readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112488781A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113111182A (en) * | 2021-04-15 | 2021-07-13 | 北京沃东天骏信息技术有限公司 | Information recommendation method and device and computer-readable storage medium |
| CN113592530A (en) * | 2021-06-24 | 2021-11-02 | 青岛海尔科技有限公司 | Recommendation method and device for service provider and electronic equipment |
| CN114969486A (en) * | 2022-08-02 | 2022-08-30 | 平安科技（深圳）有限公司 | Corpus recommendation method, apparatus, device and storage medium |
| CN113658684B (en) * | 2021-08-11 | 2024-05-31 | 挂号网（杭州）科技有限公司 | Consultation result generation method and device, electronic equipment and storage medium |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104317945A (en) * | 2014-10-31 | 2015-01-28 | 亚信科技（南京）有限公司 | E-commerce website commodity recommending method on basis of search behaviors |
| CN107908662A (en) * | 2017-10-18 | 2018-04-13 | 北京京东尚科信息技术有限公司 | The implementation method and realization device of search system |
| CN110751537A (en) * | 2019-09-29 | 2020-02-04 | 深圳市云积分科技有限公司 | Online and offline intelligent marketing method and device |
| CN111507788A (en) * | 2019-01-31 | 2020-08-07 | 阿里巴巴集团控股有限公司 | Data recommendation method and device, storage medium and processor |
| CN111881359A (en) * | 2020-08-04 | 2020-11-03 | 携程计算机技术（上海）有限公司 | Sorting method, system, equipment and storage medium in internet information retrieval |
| CN111899068A (en) * | 2020-07-07 | 2020-11-06 | 拉扎斯网络科技（上海）有限公司 | Commodity shopping guide method and device, storage medium and computer equipment |
-
2020
- 2020-11-10 CN CN202011249569.3A patent/CN112488781A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104317945A (en) * | 2014-10-31 | 2015-01-28 | 亚信科技（南京）有限公司 | E-commerce website commodity recommending method on basis of search behaviors |
| CN107908662A (en) * | 2017-10-18 | 2018-04-13 | 北京京东尚科信息技术有限公司 | The implementation method and realization device of search system |
| CN111507788A (en) * | 2019-01-31 | 2020-08-07 | 阿里巴巴集团控股有限公司 | Data recommendation method and device, storage medium and processor |
| CN110751537A (en) * | 2019-09-29 | 2020-02-04 | 深圳市云积分科技有限公司 | Online and offline intelligent marketing method and device |
| CN111899068A (en) * | 2020-07-07 | 2020-11-06 | 拉扎斯网络科技（上海）有限公司 | Commodity shopping guide method and device, storage medium and computer equipment |
| CN111881359A (en) * | 2020-08-04 | 2020-11-03 | 携程计算机技术（上海）有限公司 | Sorting method, system, equipment and storage medium in internet information retrieval |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113111182A (en) * | 2021-04-15 | 2021-07-13 | 北京沃东天骏信息技术有限公司 | Information recommendation method and device and computer-readable storage medium |
| WO2022218181A1 (en) * | 2021-04-15 | 2022-10-20 | 北京沃东天骏信息技术有限公司 | Information recommendation method and device and computer storable medium |
| CN113592530A (en) * | 2021-06-24 | 2021-11-02 | 青岛海尔科技有限公司 | Recommendation method and device for service provider and electronic equipment |
| CN113592530B (en) * | 2021-06-24 | 2023-08-22 | 青岛海尔科技有限公司 | Recommendation method and device for service provider and electronic equipment |
| CN113658684B (en) * | 2021-08-11 | 2024-05-31 | 挂号网（杭州）科技有限公司 | Consultation result generation method and device, electronic equipment and storage medium |
| CN114969486A (en) * | 2022-08-02 | 2022-08-30 | 平安科技（深圳）有限公司 | Corpus recommendation method, apparatus, device and storage medium |
| CN114969486B (en) * | 2022-08-02 | 2022-11-04 | 平安科技（深圳）有限公司 | Corpus recommendation method, apparatus, device and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111784455B (en) | Article recommendation method and recommendation equipment | |
| CN107424043B (en) | Product recommendation method and device and electronic equipment | |
| CN108629665B (en) | Personalized commodity recommendation method and system | |
| US8924314B2 (en) | Search result ranking using machine learning | |
| CN110020094B (en) | Display method and related device for search results | |
| US9727616B2 (en) | Systems and methods for predicting sales of item listings | |
| JP6177871B2 (en) | Disclosure of product information | |
| CN110532479A (en) | A kind of information recommendation method, device and equipment | |
| US20160124996A1 (en) | Image ranking based on attribute correlation | |
| CN107209762A (en) | Visual interactive formula is searched for | |
| CN112488781A (en) | Search recommendation method and device, electronic equipment and readable storage medium | |
| JP5859606B2 (en) | Ad source and keyword set adaptation in online commerce platforms | |
| US20140172821A1 (en) | Generating filters for refining search results | |
| CN109635198B (en) | Method, device, medium and electronic equipment for presenting user search results on commodity display platform | |
| CN110175895B (en) | Article recommendation method and device | |
| EP2842060A1 (en) | Recommending keywords | |
| US11301540B1 (en) | Refined search query results through external content aggregation and application | |
| EP2438509A1 (en) | System and method for learning user genres and styles and matching products to user preferences | |
| US9330071B1 (en) | Tag merging | |
| US11682060B2 (en) | Methods and apparatuses for providing search results using embedding-based retrieval | |
| CN111967924A (en) | Commodity recommendation method, commodity recommendation device, computer device, and medium | |
| CN113971599A (en) | Advertisement putting and selecting method and device, equipment, medium and product thereof | |
| KR20220001618A (en) | Method, Apparatus and System for Recommendation in Groups Using Bigdata | |
| KR20220001617A (en) | Method, Apparatus and System for Item Recommendation Using Consumer Bigdata | |
| CN111654714A (en) | Information processing method, device, electronic equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |