Information search method and apparatus, and device and storage medium

Abstract

Provided in the embodiments of the present disclosure are an information search method and apparatus, and a device and a storage medium. The method comprises: on the basis of a received search request, determining a target tag corresponding to the search request; on the basis of the comparison between the target tag and a tag set associated with a group of points of interest, determining from the group of points of interest at least one point of interest that matches the target tag, wherein the tag set is generated at least on the basis of user interaction information associated with the group of points of interest; and on the basis of the at least one point of interest, determining a search result for the search request. The present disclosure can effectively improve the efficiency and accuracy of information search.

Description

Information search methods, devices, equipment and storage media Technical Field

[0001] The exemplary embodiments disclosed herein generally relate to the field of computers, and particularly to methods, apparatus, devices, and computer-readable storage media for information retrieval. Background Technology

[0002] With the development of computer technology, various forms of electronic devices have greatly enriched people's daily lives. For example, people can use electronic devices to search for information, such as entertainment venues and restaurants. How to improve the efficiency of information retrieval is a key concern. Summary of the Invention

[0003] In a first aspect of this disclosure, an information retrieval method is provided. The method includes: determining a target tag corresponding to a received search request; determining at least one point of interest matching the target tag from the set of points of interest based on a comparison between the target tag and a set of tags associated with a set of points of interest, wherein the set of tags is generated at least based on user interaction information associated with the set of points of interest; and determining a search result for the search request based on the at least one point of interest.

[0004] In a second aspect of this disclosure, an apparatus for information retrieval is provided. The apparatus includes: a first determining module configured to determine a target tag corresponding to a received search request; a second determining module configured to determine at least one point of interest matching the target tag from a set of points of interest based on a comparison of the target tag and a set of tags associated with a set of points of interest, wherein the set of tags is generated at least based on user interaction information associated with the set of points of interest; and a third determining module configured to determine the search results of the search request based on the at least one point of interest.

[0005] In a third aspect of this disclosure, an electronic device is provided. The device includes at least one processing unit; and at least one memory coupled to the at least one processing unit and storing instructions for execution by the at least one processing unit. When executed by the at least one processing unit, the instructions cause the device to perform the method of the first aspect.

[0006] In a fourth aspect of this disclosure, a computer-readable storage medium is provided. The computer-readable storage medium stores a computer program that can be executed by a processor to implement the method of the first aspect.

[0007] In a fifth aspect of this disclosure, a computer program product is provided. The computer program product includes computer-executable instructions that, when executed by a processor, implement the method according to a first aspect of this disclosure.

[0008] It should be understood that the content described in this content section is not intended to limit the key or essential features of the embodiments of this disclosure, nor is it intended to restrict the scope of this disclosure. Other features of this disclosure will become readily apparent from the following description. Attached Figure Description

[0009] The above and other features, advantages, and aspects of the embodiments of this disclosure will become more apparent from the accompanying drawings and the following detailed description. In the drawings, the same or similar reference numerals denote the same or similar elements, wherein:

[0010] Figure 1 shows a schematic diagram of an example environment in which embodiments of the present disclosure can be implemented;

[0011] Figure 2 illustrates a flowchart of an information search process according to some embodiments of the present disclosure;

[0012] Figure 3 shows a flowchart of a process for recalling points of interest based on RTag according to some embodiments of the present disclosure;

[0013] Figure 4 shows a flowchart of the label generation process according to some embodiments of the present disclosure;

[0014] Figure 5 shows a flowchart of the point-of-interest recall process according to some embodiments of the present disclosure;

[0015] Figure 6 shows a schematic structural block diagram of an apparatus for information retrieval according to certain embodiments of the present disclosure;

[0016] Figure 7 shows a block diagram of an electronic device capable of implementing several embodiments of the present disclosure. Detailed Implementation

[0017] Embodiments of this disclosure will now be described in more detail with reference to the accompanying drawings. While some embodiments of this disclosure are shown in the drawings, it should be understood that this disclosure can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of this disclosure. It should be understood that the accompanying drawings and embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of protection of this disclosure.

[0018] It should be noted that the headings of any section / subsection provided herein are not limiting. Various embodiments are described throughout this document, and embodiments of any type may be included under any section / subsection. Furthermore, embodiments described in any section / subsection may be combined in any way with any other embodiments described in the same section / subsection and / or different sections / subsections.

[0019] In the description of embodiments of this disclosure, the term "comprising" and similar terms should be understood as open-ended inclusion, i.e., "including but not limited to". The term "based on" should be understood as "at least partially based on". The term "one embodiment" or "the embodiment" should be understood as "at least one embodiment". The term "some embodiments" should be understood as "at least some embodiments". Other explicit and implicit definitions may also be included below. The terms "first", "second", etc., may refer to different or the same objects. Other explicit and implicit definitions may also be included below.

[0020] The embodiments of this disclosure may involve user data, data acquisition, and / or use. All of these aspects comply with applicable laws, regulations, and relevant provisions. In the embodiments of this disclosure, all data collection, acquisition, processing, manipulation, forwarding, and use are conducted with the user's knowledge and confirmation. Accordingly, in implementing the embodiments of this disclosure, the type, scope of use, and usage scenarios of any data or information that may be involved should be communicated to the user and their authorization obtained in accordance with relevant laws and regulations through appropriate means. The specific methods of notification and / or authorization may vary depending on the actual situation and application scenario, and the scope of this disclosure is not limited in this respect.

[0021] In this specification and the embodiments, any processing of personal information will be carried out only under the premise of legality (such as obtaining the consent of the personal information subject, or being necessary for the performance of a contract), and will only be carried out within the scope stipulated or agreed upon. A user's refusal to process personal information other than that necessary for basic functions will not affect the user's use of basic functions.

[0022] Embodiments of this disclosure propose an information search scheme. According to this scheme, based on a received search request, a target tag corresponding to the search request is determined; and based on a comparison between the target tag and a set of tags associated with a set of points of interest, at least one point of interest matching the target tag is determined from the set of points of interest, wherein the set of tags is generated at least based on user interaction information associated with the set of points of interest; and based on the at least one point of interest, a search result for the search request is determined.

[0023] Based on this approach, embodiments of this disclosure can locate points of interest related to a search request by comparing the target tag corresponding to the search request with a set of tags associated with a set of points of interest, effectively improving the efficiency and accuracy of information retrieval. Furthermore, the tag set of this disclosure is generated based on user interaction information, which can accurately capture user preference information and improve the quality of search results.

[0024] Example Environment

[0025] Figure 1 illustrates a schematic diagram of an example environment 100 in which embodiments of the present disclosure can be implemented. As shown in Figure 1, the example environment 100 may include an electronic device 110.

[0026] In this example environment 100, electronic device 110 can determine the search results corresponding to a received search request. The search request can be any appropriate request, such as a search request related to restaurants, a search request related to entertainment venues, and so on.

[0027] Electronic device 110 can be any type of mobile terminal, fixed terminal, or portable terminal, including mobile phones, desktop computers, laptop computers, notebook computers, netbook computers, tablet computers, media computers, multimedia tablets, handheld computers, portable gaming terminals, VR / AR devices, personal communication system (PCS) devices, personal navigation devices, personal digital assistants (PDAs), audio / video players, digital cameras / camcorders, positioning devices, television receivers, radio receivers, e-book devices, gaming devices, or any combination thereof, including accessories and peripherals of these devices or any combination thereof. In some embodiments, electronic device 110 can also support any type of interface for the target user (such as "wearable" circuitry).

[0028] Electronic device 110 can also be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks, and big data and artificial intelligence platforms. Electronic device 110 may include, for example, computing systems / servers, such as mainframes, edge computing nodes, computing devices in cloud environments, and so on.

[0029] It should be understood that the structure and function of the various elements in environment 100 are described for illustrative purposes only and do not imply any limitation on the scope of this disclosure.

[0030] The following description will continue with reference to the accompanying drawings, which will provide some exemplary embodiments of this disclosure.

[0031] Example process

[0032] Figure 2 shows a flowchart of an information search process 200 according to some embodiments of the present disclosure. Process 200 can be implemented at electronic device 110. Process 200 is described below with reference to Figure 1.

[0033] In box 210, electronic device 110 determines the target tag corresponding to the search request based on the received search request.

[0034] In some embodiments, a search request can be user-inputted data containing any appropriate keywords, descriptive phrases, etc., which can be related to the service, personal needs, etc. For example, a search request could be "restaurants for couples", "coffee shops suitable for pregnant women", "find a fun entertainment venue in City A", etc.

[0035] In some embodiments, the electronic device 110 can utilize a tag extraction model to determine target tags corresponding to the search request. Specifically, the electronic device 110 can use a tag extraction model to extract keywords corresponding to the search request and map them to corresponding target tags. Target tags are information that reflects the user's search intent and is associated with points of interest. Points of interest can be any appropriate location that can provide specific services or functions, such as tourist attractions, libraries, restaurants, supermarkets, etc. For example, if the search request is "restaurants for couples' dates," then the corresponding target tag can be "restaurants for couples," indicating that the user expects to search for restaurants suitable for couples' dates.

[0036] In box 220, electronic device 110 determines at least one point of interest from a set of points of interest that matches the target label based on a comparison between the target label and a set of labels associated with a set of points of interest, wherein the set of labels is generated based at least on user interaction information associated with the set of points of interest.

[0037] To improve the accuracy of tag matching and thus the efficiency of information retrieval, in some embodiments, the electronic device 110 may replace the target tag with the first tag in response to a similarity between the target tag and a first tag in the tag set being greater than a first threshold but less than a second threshold. The first threshold and the second threshold can be set as needed, wherein the first threshold is less than the second threshold.

[0038] In some embodiments, the first tag may include the name corresponding to the point of interest. For example, if the target tag corresponding to the search request is "XXX's restaurant", and the first tag in the set of tags corresponding to this set of points of interest is only "XXX restaurant", then the electronic device 110 can replace the target tag "XXX's restaurant" with "XXX restaurant".

[0039] In other embodiments, the first tag may also include the category corresponding to the point of interest. For example, if the target tag corresponding to the search request is "Korea", and the first tag in the set of tags corresponding to this set of points of interest is only "Korean style" and "Korean cuisine", then the electronic device 110 can replace the target tag "Korea" with "Korean style" or "Korean cuisine".

[0040] In other embodiments, the first tag may also include the infrastructure corresponding to the point of interest. The infrastructure can be any suitable facility; for example, if the point of interest is a restaurant, the infrastructure could be private rooms, etc. For instance, if the target tag corresponding to the search request is "private rooms available," and the first tag in the set of tags corresponding to this set of points of interest is only "private rooms for fewer than 10 people," "private rooms for 10-20 people," or "private rooms for more than 20 people," then the electronic device 110 can replace the target tag "private rooms available" with at least one of "private rooms for fewer than 10 people," "private rooms for 10-20 people," or "private rooms for more than 20 people."

[0041] In some embodiments, the target tag may include multiple tags (i.e., the target tag may include a second tag and a third tag). After recalling multiple sets of points of interest based on each tag, the intersection between these multiple sets of points of interest may be zero or the number of points of interest included in the intersection may be too small, resulting in insufficient richness of search results available to the user, thereby reducing the user experience. Therefore, the electronic device 110 can determine at least one set of points of interest from each set of points of interest as at least one point of interest matching the target tag. It should be noted that the second tag and the third tag do not limit the number of target tags; they only indicate that the target tag includes different tags.

[0042] Specifically, the electronic device 110 can determine a first set of candidate interest points from a set of interest points based on a comparison between the second tag and the tag set. Further, the electronic device 110 can determine a second set of candidate interest points from a set of interest points based on a comparison between the third tag and the tag set. Further, the electronic device 110 can determine at least one interest point matching the target tag based on the first and second sets of candidate interest points. Specifically, the electronic device can determine the at least one interest point from the first set of candidate interest points and the second set of candidate interest points based on the types of the second and third tags. The types can include multi-person types and non-multi-person types, where a multi-person type indicates that the tag is associated with multiple people, and a non-multi-person type indicates that it is not associated with multiple people. In some embodiments, multi-person type tags can include, but are not limited to, "couple's date," "friends' gathering," "girls' party," "business dinner," "birthday party," "family gathering," etc.

[0043] As an example, an electronic device may, in response to the fact that both the second and third tags are of the multi-person type, determine either the first group of candidate points of interest or the second group of candidate points of interest as at least one point of interest that matches the target tag, or determine the points of interest included in the union of the first group of candidate points of interest or the second group of candidate points of interest as at least one point of interest that matches the target tag.

[0044] As another example, an electronic device may, in response to the fact that both the second and third tags are of the multi-person type and the number of interest points included in the intersection of the first group of candidate interest points and the second group of candidate interest points is less than a threshold, determine the first group of candidate interest points or the second group of candidate interest points as at least one interest point that matches the target tag.

[0045] As another example, an electronic device may, in response to the fact that the types of the first label and the third label are not both multi-person types, determine the interest points included in the intersection of the first group of candidate interest points and the second group of candidate interest points as at least one interest point that matches the target label.

[0046] Figure 3 shows a flowchart of a process for recalling points of interest based on target tags (RTags) according to some embodiments of the present disclosure, and will now be described with reference to Figure 3.

[0047] In box 310, electronic device 110 determines whether RTag has an alias.

[0048] As an example, an electronic device determines whether an RTag has a tag with a similar name in the tag set. For instance, if the target tag is "XX restaurant" and the tag set contains the tag "XX's restaurant", then it is determined that the RTag has an alias "XX's restaurant".

[0049] In some embodiments, electronic device 110 may perform the operation of block 320 in response to determining that RTag has an alias. Electronic device 110 may perform the operation of block 330 in response to determining that RTag does not have an alias.

[0050] In box 320, electronic device 110 replaces the RTag name.

[0051] In some embodiments, the electronic device can replace the RTag with an alias corresponding to the RTag. As an example, the electronic device can replace "XX Restaurant" with "XX's Restaurant".

[0052] In box 330, electronic device 110 determines whether a product category needs to be changed.

[0053] Electronic devices can determine whether an RTag has a corresponding similar category in the tag set. For example, if the target tag for a search request is "Korea," and the tag set includes "Korean cuisine," then it is determined that a category change is needed.

[0054] In some embodiments, electronic device 110 may perform the operation of block 340 in response to the need for category conversion. Electronic device 110 may perform the operation of block 350 in response to the need for category conversion.

[0055] In box 340, replace RTag with category filter for electronic devices 110.

[0056] As an example, electronic device 110 can replace the target label "Korea" with "Korean cuisine".

[0057] In box 350, electronic device 110 determines whether RTag is infrastructure.

[0058] As an example, taking RTag as a compartment, electronic devices can identify RTag as infrastructure.

[0059] In some embodiments, electronic device 110 may perform the operation of block 360 in response to determining that RTag is infrastructure. Electronic device 110 may perform the operation of block 370 in response to determining that RTag is not infrastructure.

[0060] In box 360, electronic devices 110 replace RTag with infrastructure filter.

[0061] As an example, if the tag set contains "private room for less than 10 people", "private room for 10-20 people", and "private room for more than 20 people", then the electronic device 110 can replace RTag with at least one of "private room for less than 10 people", "private room for 10-20 people", and "private room for more than 20 people".

[0062] In box 370, electronic device 110 determines whether RTag is a multi-person RTag.

[0063] In some embodiments, the first and second tags included in the target tag of the multi-person RTag are both of the type multi-person type. The multi-person type tags may include, but are not limited to, "couple date", "friends gathering", "girls' gathering", "business banquet", "birthday party", "family gathering", etc. As an example, when the RTag includes "couple date" and "friends gathering", the corresponding RTag is a multi-person RTag.

[0064] In some embodiments, electronic device 110 may perform the operation of block 380 in response to determining that the RTag is a multi-user RTag. Electronic device 110 may perform the operation of block 390 in response to determining that the RTag is not a multi-user RTag.

[0065] In box 380, the electronic devices 110 are in union with the RTag filter.

[0066] As an example, when the RTag includes "couple's date" and "friends' gathering", the electronic device can determine a first set of candidate points of interest corresponding to "couple's date" and a second set of candidate points of interest corresponding to "friends' gathering". Further, the electronic device can determine the union of the first and second sets of candidate points of interest. Further still, the electronic device can determine the points of interest included in the union as at least one point of interest matching the RTag.

[0067] In box 390, the intersection of electronic devices 110 and the RTag filter is used.

[0068] As an example, when the RTag includes "couple dating" and "fast food," the electronic device can determine a first set of candidate points of interest corresponding to "couple dating" and a third set of candidate points of interest corresponding to "fast food." Furthermore, the electronic device can determine the intersection of the first and third sets of candidate points of interest. Further, the electronic device can identify the points of interest included in the intersection as at least one point of interest matching the RTag.

[0069] For ease of description, we will take the generation of a set of tags associated with points of interest by electronic device 110 as an example to illustrate the tag set generation process. It should be noted that the tag set can also be generated by devices other than electronic device 110, such as a server, which will not be elaborated here.

[0070] In some embodiments, the electronic device 110 can generate descriptive text about a set of points of interest based on user interaction information. The user interaction information can be any suitable information associated with the points of interest, such as user comments on the points of interest, the number of times the user clicked on the points of interest, the number of times the user searched for the points of interest, etc. The descriptive text can be a detailed description of the points of interest, including, but not limited to, explanatory information about the points of interest in various dimensions. As an example, if the point of interest is a restaurant, the descriptive text can include reviews of dish A at this restaurant, descriptions of highly rated dishes, etc. As an example, the electronic device 110 can provide user interaction information to a first model to generate multiple data chunks. The first model can be any suitable machine learning model used to segment the data to obtain data chunks. Further, the electronic device 110 can provide multiple data chunks to a second model to generate descriptive text about a set of points of interest. The second model can be any suitable machine learning model used to summarize and analyze the data chunks to generate a detailed description of the points of interest.

[0071] In some embodiments, the descriptive text may also be determined based on media content associated with a set of points of interest. Specifically, the descriptive text may also be determined based on the Automatic Speech Recognition (ASR) results corresponding to the media content associated with the set of points of interest. ASR is text information associated with a point of interest recognized based on video data and / or audio data associated with that point. In some embodiments, the descriptive text may also be determined based on other data sources associated with the points of interest. As an example, the electronic device 110 may provide the first model with the ASR corresponding to the media content associated with the set of points of interest, as well as other data sources, to generate multiple data blocks. Further, the electronic device 110 may provide the second model with multiple data blocks to generate descriptive text about the set of points of interest.

[0072] Furthermore, the electronic device 110 can determine a set of tags associated with a set of points of interest based on descriptive text and metadata of that set. Metadata can characterize the basic features and attributes corresponding to a point of interest, such as the services it provides, its location information, and its environmental information. For example, if a point of interest is a restaurant, its metadata could include information about the restaurant's menu, its environmental information, its tag information, its location information, and so on.

[0073] As an example, electronic device 110 can provide a third model with descriptive text and metadata for a set of points of interest to generate multiple candidate labels associated with the set of points of interest. The third model can be any suitable machine learning model. Specifically, electronic device 110 can assemble the descriptive text and metadata of this set of points of interest to obtain assembled data corresponding to this set of points of interest. Further, electronic device 110 can provide the assembled data corresponding to this set of points of interest to the third model to generate multiple candidate labels associated with the set of points of interest.

[0074] As another example, the electronic device 110 can also utilize a predefined rule engine to determine multiple candidate tags associated with a set of points of interest (POIs) based on the descriptive text corresponding to a set of POIs and the metadata of a set of POIs. The predefined rule engine is configured with a set of predefined tags, and for each predefined tag, there is a predefined rule. Specifically, the electronic device 110 can assemble the descriptive text of the set of POIs and the metadata of the set of POIs to obtain the assembled data corresponding to the set of POIs. Further, the electronic device 110 can determine the score of each POI for that predefined tag, based on the assembled data corresponding to that POI and the rule corresponding to that predefined tag, for each predefined tag in the set of predefined tags and each POI in the set of POIs. The higher the match between the assembled data and the rule corresponding to the predefined tag, the higher the score. For example, if a predefined tag is "spicy hot pot," and hot pot in location A is predefined as spicy, while hot pot in other locations is predefined as non-spicy, then if point of interest C offers dishes including hot pot from location A, then point of interest C can score 100 points for the predefined tag "spicy hot pot." Furthermore, the electronic device 110 can determine a predetermined number of tags with the highest scores as candidate tags.

[0075] Furthermore, the electronic device 110 can determine a set of tags associated with a group of points of interest based on multiple candidate tags. As an example, in addition to generating multiple candidate tags, the third model can also generate first evaluation information corresponding to these multiple candidate tags. In response to the first evaluation information of the target candidate tag satisfying a preset condition, the electronic device 110 can add the target candidate tag to the tag set, wherein the target candidate tag is at least one tag among the multiple candidate tags.

[0076] Figure 4 shows a flowchart of the label generation process according to some embodiments of the present disclosure, and will now be described with reference to Figure 4.

[0077] In box 410, electronic device 110 can use a language model to generate chunk data based on point-of-interest comments, ASR, and other data sources.

[0078] In some embodiments, ASR is text information associated with a point of interest (POI) identified based on video and / or audio data associated with the POI. POI comments can be user-generated comments about the POI. The language model can be any suitable machine learning model, and chunk data can also be referred to as data partitioning.

[0079] Specifically, electronic device 110 can input point of interest comments, ASR, and other data sources into the language model to obtain chunk data output by the language model.

[0080] In box 420, electronic device 110 uses a language model to generate report data.

[0081] As an example, electronic device 110 can input chunk data into a language model to obtain report data generated by the language model. The language model in box 420 can be the same as or different from the language model in box 410.

[0082] In box 430, electronic device 110 can assemble metadata of points of interest and report data to obtain an assembled data source.

[0083] In some embodiments, metadata can characterize the basic features and attributes corresponding to a point of interest, such as the services that the point of interest can provide, its location information, its environmental information, etc. The assembled data source can include reference information for various dimensions of the point of interest.

[0084] In box 440, electronic device 110 obtains the tag (RTag) configuration corresponding to the point of interest.

[0085] As an example, electronic device 110 can determine the RTag configuration corresponding to a point of interest (POI) based on an engineering tagging method. Specifically, electronic device 110 can utilize a rule engine to determine the degree of matching between the assembled data source corresponding to the POI and the rules corresponding to the target predefined tag. Further, in response to the assembled data source corresponding to the POI determining that the degree of matching between the assembled data source and the rules corresponding to the target predefined tag is greater than a threshold, electronic device 110 can identify the target predefined tag as the RTag corresponding to the POI and determine the score corresponding to the RTag, where the score is correlated with the degree of matching.

[0086] As an example, electronic device 110 can determine the RTag configuration corresponding to a point of interest based on a model tagging method. Specifically, electronic device 110 can input the assembled data source corresponding to the point of interest into the model to obtain the RTag corresponding to the point of interest and the score corresponding to the RTag output by the model. In some embodiments, the model can also output the target reason for the RTag corresponding to the point of interest.

[0087] In box 450, electronic device 110 obtains the RTag corresponding to the points of interest that meet the criteria.

[0088] As an example, electronic device 110 can identify RTags with scores greater than a threshold as RTags corresponding to points of interest that meet the criteria.

[0089] In box 460, electronic device 110 stores the RTag corresponding to the point of interest in the database.

[0090] Returning to Figure 2, in box 230, electronic device 110 determines the search results for a search request based on at least one point of interest.

[0091] In some embodiments, the electronic device 110 can determine a set of candidate points of interest (POIs) that match a search request based on at least one POI. As an example, the electronic device 110 can add all at least one POI to the candidate POI set. As another example, the electronic device 110 can filter the at least one POI based on predetermined filtering rules, adding unfiltered POIs to the candidate POI set. The filtering rules can be any suitable rules, such as filtering POIs that are not currently in business hours.

[0092] In some embodiments, the candidate point of interest set further includes a set of additional points of interest determined based on at least one query term, which is determined based on a search request. In some embodiments, the query term is information other than the target label.

[0093] In some embodiments, to improve the accuracy of information retrieval, in addition to determining the at least one point of interest based on the target tag corresponding to the search request, additional points of interest can be determined based on information from other dimensions associated with the search request. This other dimension information may include at least one of the following: average per capita, category, brand, distance, rating, etc.

[0094] Specifically, electronic device 110 can utilize multiple models to determine parameters for other dimensions corresponding to the search request. Furthermore, electronic device 110 can recall or determine additional points of interest based on the target label and at least one of the other dimensions.

[0095] In some embodiments, different recall strategies can be set for different query scenarios. For example, for map query scenarios, points of interest can be sorted from highest to lowest rating, and a predetermined number of points of interest at the top of the sorted ranking can be recalled to determine additional points of interest. The predetermined number can be set according to needs, for example, it can be 150. For homepage query scenarios, points of interest with a distance less than a threshold can be sorted from nearest to farthest, and a predetermined number of points of interest at the top of the sorted ranking can be recalled to determine additional points of interest. The predetermined number can be set according to needs, for example, it can be 150. The threshold can be any appropriate threshold, for example, it can be 20km.

[0096] For ease of description, the parameters of each dimension determined based on the search request are referred to as a set of parameters (which may include at least one dimension of the target label and / or other dimension parameters). In some embodiments, the electronic device 110 may perform pre-parameter processing based on this set of parameters, that is, convert this set of parameters into parameters that are convenient for point-of-interest retrieval.

[0097] In some embodiments, the electronic device 110 can determine multiple filtering parameters corresponding to this set of parameters, which are used to characterize the corresponding recall rules. In some embodiments, the filtering parameters corresponding to these multiple dimensions may include geographic location, included items, excluded items, filtering conditions, non-empty conditions, sorting conditions, regional recall, group purchase recall, etc.

[0098] In some embodiments, geolocation represents the location requirement for recalling points of interest. Geographic location includes a center point and city restrictions. For example, if a user's search request is "good restaurants in Beijing", the corresponding filter parameters may include geographic location, and the specific recall location is restaurants located in Beijing.

[0099] In some embodiments, the inclusion term represents the requirements that the recalled points of interest must meet. The inclusion term may include tags (RTag), keywords, recommended dishes, point of interest identifiers, categories, and brands, where the point of interest identifier can be the name of the point of interest. For example, if a user's search request is "find a restaurant that serves roast beef," the corresponding filter parameters may include the inclusion term, specifically recalling restaurants that offer "roast beef."

[0100] In some embodiments, exclusions characterize which requirements the recalled points of interest cannot meet. Exclusions may include RTag, keywords, recommended dishes, point of interest identifiers, categories, and brands. For example, if a user's search request is "find a non-fast food restaurant," the corresponding filter parameters may include exclusions, specifically recalling restaurants that do not offer fast food services.

[0101] In some embodiments, filter criteria represent the conditions upon which points of interest are retrieved. Filter criteria may include, but are not limited to, ratings, distance, opening hours, average price per person, and whether videos are included. For example, if a user's search request is "find a restaurant with a rating higher than 4.5", the corresponding filter parameters may include filter criteria, specifically retrieving restaurants with ratings higher than 4.5.

[0102] In some embodiments, ranking criteria represent the conditions used to rank points of interest in order to recall those ranked higher. Ranking criteria may include, but are not limited to, proximity priority, low price priority, high price priority, positive reviews priority, etc. For example, if a user's search request is "Please help me find a restaurant close to my current location," the corresponding filtering parameters may include ranking criteria, specifically recalling restaurants that are relatively close.

[0103] In some embodiments, the region recall representation is based on which region dimension to recall points of interest. Specifically, the region dimension can include, but is not limited to, city, region, business district, shopping mall, etc. For example, if a user's search request is "Please help me find a good restaurant in shopping mall A", then the corresponding filtering parameters can include region recall, specifically recalling restaurants located in shopping mall A.

[0104] In some embodiments, group-buying recall represents the points of interest that can provide group-buying services. For example, if a user's search request is "3-person package", the corresponding filter parameters can include group-buying recall, and the specific recall can provide the points of interest for a 3-person group-buying package.

[0105] Furthermore, the electronic device 110 can determine additional points of interest that match these multiple dimensions of filtering parameters based on these multiple dimensions of filtering parameters.

[0106] Furthermore, the electronic device 110 can determine a set of target interest points from the candidate interest point set based on second evaluation information associated with the candidate interest point set, as search results. The second evaluation information can be a score, which can characterize the quality of the interest point, its convenience relative to the user's current location, etc. As an example, the electronic device 110 can determine a predetermined number of candidate interest points with the highest second evaluation information in the candidate interest point set as a set of target interest points. As another example, the electronic device 110 can determine a set of target interest points as candidate interest points whose second evaluation information meets predetermined evaluation criteria.

[0107] In some embodiments, the second evaluation information may be determined based on at least one of the following: the relevance (Es score) between the search request and the target description information of the point of interest, the degree of matching between the search request and the point of interest, business parameters associated with the point of interest, and the search scenario associated with the search request. Taking a restaurant as an example, the target description information of the point of interest may include the restaurant's category, brand, name, and recommended dishes. The degree of matching between the search request and the point of interest may be the text matching degree between the search request and the keywords corresponding to the point of interest, and the keywords corresponding to the point of interest may be the name, brand, category, recommended dishes, etc., of the point of interest. Business parameters include distance, rating, rankings, etc.

[0108] Specifically, in some embodiments, the electronic device 110 can determine the first score (Es score) for each interest point in the candidate interest point set based on the product of the relevance of the search request and the target description information of the candidate interest point set and the corresponding weights.

[0109] In some embodiments, the electronic device 110 can perform text matching based on a set of keywords corresponding to a search request and a point of interest, and determine a second score corresponding to each point of interest in the candidate point of interest set according to the matching results. Specifically, for each keyword in this set of keywords, the electronic device 110 can determine the sub-score corresponding to that keyword in the search request and the point of interest.

[0110] Furthermore, the electronic device 110 can determine a first weight sum based on the search request, the sub-score of each keyword, and the weight corresponding to each keyword. This first weight sum is also the second score.

[0111] In some embodiments, the electronic device 110 can determine multiple sub-scores for each interest point in the candidate interest point set based on multiple service parameters associated with the candidate interest point set, wherein one service parameter corresponds to one sub-score.

[0112] Furthermore, the electronic device 110 can determine a second weight sum based on the multiple sub-scores for each business parameter in the search request and the weight corresponding to each business parameter. This second weight sum is also the third score.

[0113] Furthermore, the electronic device 110 can determine a target score (second evaluation information) based on the first score, the second score, and the third score. Specifically, the electronic device can determine the target score as the sum of the first score, the second score, and the third score.

[0114] In some embodiments, if the electronic device 110 does not identify any entity from the search request, the entity may include tags, other dimension information such as the name of the point of interest, brand category, service, etc. For example, the search request is "I want to find a restaurant". To improve the accuracy of information search, the electronic device 110 can determine the target point of interest as the search result corresponding to the search request based on the search request and predetermined recall rules. Specifically, the electronic device 110 can determine which demand in a set of demands the search request matches. Further, in response to determining that the search request matches a target demand in a set of demands, the electronic device 110 can use the point of interest associated with the target demand as the target point of interest. For example, if the search request matches a beverage demand, such as the search request being "delicious", the electronic device 110 can restrict the category corresponding to the point of interest, that is, determine beverage shops as the target point of interest. As another example, if the search request matches a demand other than beverage demand, such as the search request being "delicious" or "gourmet food", the electronic device 110 can restrict the category corresponding to the point of interest, that is, determine fast food restaurants as the target point of interest. For example, if a search request matches a request other than for beverages or fast food, such as for "restaurant" or "food", then electronic device 110 can restrict the category corresponding to the point of interest, that is, determine the restaurant as the target point of interest.

[0115] In some embodiments, if the electronic device 110 does not determine the tag, the name of the point of interest, or the brand of the point of interest from the search request, and the current search scenario is a non-map scenario, then the point of interest with a score greater than a threshold can be identified as the target point of interest. The threshold can be set according to needs, for example, it can be 4.

[0116] Figure 5 shows a flowchart of the point-of-interest recall process according to some embodiments of the present disclosure, and will now be described with reference to Figure 5.

[0117] In box 510, electronic device 110 uses the parameter processing model to extract parameters.

[0118] As an example, electronic device 110 can extract information from multiple dimensions such as target tags, categories, brands, and average cost per person corresponding to a search request (query) based on multiple parameter extraction models.

[0119] In box 520, electronic device 110 performs parameter processing.

[0120] As an example, electronic device 110 can determine whether the search scenario associated with the search request is a map search scenario or a homepage search scenario. For instance, in response to determining that the search scenario is a map search scenario, electronic device 110 can determine that the number of points of interest to be recalled is 150, and the points of interest to be recalled are based on a default order of priority based on positive reviews. As another example, electronic device 110 can, in response to determining that the search scenario is a homepage search scenario, determine that the number of points of interest to be recalled is 150, and the points of interest to be recalled are based on a default order of priority based on distance, with a default maximum distance of 20km between the recalled points of interest and the user's current location.

[0121] In box 530, electronic device 110 performs pre-processing of policies.

[0122] As an example, electronic device 110 can convert the extracted information from multiple dimensions and the results of parameter processing into a domain-specific language (DSL) that facilitates subsequent point-of-interest retrieval.

[0123] In box 540, electronic device 110 performs point of interest recall processing.

[0124] As an example, electronic device 110 can recall the first set of points of interest based on DSL.

[0125] In frame 550, electronic device 110 performs post-filtering.

[0126] As an example, electronic devices can filter some points of interest based on predetermined filtering rules, such as filtering out points of interest that are not in business hours at the current time, in order to obtain a second set of points of interest.

[0127] In box 560, electronic device 110 rearranges points of interest.

[0128] As an example, the electronic device 110 can determine the score corresponding to the second group of interest points based on factors such as the relevance of the query to the target description information of the second group of interest points, the text matching degree of the query to the predetermined keywords of the second group of interest points, and the business parameters of the second group of interest points, and then rearrange the second group of interest points in descending order of score.

[0129] In box 570, electronic device 110 returns the search results.

[0130] As an example, electronic device 110 can present a predetermined number of points of interest with the highest scores in the second group of points of interest to the user as search results.

[0131] Based on this approach, embodiments of this disclosure can locate points of interest related to a search request by comparing the target tag corresponding to the search request with a set of tags associated with a set of points of interest, effectively improving the efficiency and accuracy of information retrieval. Furthermore, the tag set of this disclosure is generated based on user interaction information, which can accurately capture user preference information and improve the quality of search results.

[0132] Example devices and equipment

[0133] Embodiments of this disclosure also provide corresponding apparatus for implementing the methods or processes described above. Figure 6 shows a schematic structural block diagram of an information search apparatus 600 according to certain embodiments of this disclosure. Apparatus 600 may be implemented as or included in the electronic device 110 discussed above. The various modules / components in apparatus 600 may be implemented by hardware, software, firmware, or any combination thereof.

[0134] As shown in Figure 6, the apparatus 600 includes a first determining module 610 configured to determine a target tag corresponding to a received search request; a second determining module 620 configured to determine at least one point of interest matching the target tag from a set of points of interest based on a comparison between the target tag and a set of tags associated with a set of points of interest, wherein the set of tags is generated based at least on user interaction information associated with the set of points of interest; and a third determining module 630 configured to determine the search results of the search request based on at least one point of interest. In some embodiments, the set of tags corresponding to a set of points of interest is determined based on the following process: generating descriptive text about the set of points of interest based on user interaction information; and determining the set of tags associated with the set of points of interest based on the descriptive text and metadata of the set of points of interest.

[0135] In some embodiments, generating descriptive text about a set of points of interest based on user interaction information includes: providing user interaction information to a first model to generate multiple data blocks; and providing multiple data blocks to a second model to generate descriptive text about a set of points of interest.

[0136] In some embodiments, the descriptive text is also determined based on media content associated with a set of points of interest.

[0137] In some embodiments, determining a set of tags associated with a set of points of interest based on descriptive text and metadata of a set of points of interest includes: providing descriptive text and metadata of a set of points of interest to a third model to generate multiple candidate tags associated with a set of points of interest; and determining a set of tags associated with a set of points of interest based on the multiple candidate tags.

[0138] In some embodiments, determining a set of tags associated with a set of points of interest based on multiple candidate tags includes: adding the target candidate tag to the tag set in response to a first evaluation information of the target candidate tag satisfying a preset condition.

[0139] In some embodiments, the second determining module 620 is further configured to: replace the target label with the first label in response to the similarity between the target label and the first label in the label set being greater than a first threshold but less than a second threshold; and determine at least one point of interest based on the comparison between the first label and the label set.

[0140] In some embodiments, the first tag includes at least one of the following: the name corresponding to the point of interest, the category corresponding to the point of interest, and the infrastructure corresponding to the point of interest.

[0141] In some embodiments, the target label includes a second label and a third label, and the second determining module 620 is further configured to: determine a first group of candidate interest points from a set of interest points based on a comparison of the second label with a label set; determine a second group of candidate interest points from a set of interest points based on a comparison of the third label with a label set; and determine the at least one interest point based on the first group of candidate interest points and the second group of candidate interest points.

[0142] In some embodiments, the second determining module 620 is further configured to: determine at least one interest point from the first group of candidate interest points and the second group of candidate interest points based on the types of the second label and the third label.

[0143] In some embodiments, the third determining module 630 is further configured to: determine a set of candidate points of interest that match the search request based on at least one point of interest; and determine a set of target points of interest from the set of candidate points of interest based on second evaluation information associated with the set of candidate points of interest, as search results.

[0144] In some embodiments, the candidate interest set may further include a set of additional interest points determined based on at least one query term, which is determined based on a search request.

[0145] In some embodiments, the second evaluation information associated with the candidate set of points of interest is determined based on at least one of the following: the relevance of the search request to the target description information of the points of interest; the degree of matching between the search request and the points of interest; business parameters associated with the points of interest; and the search scenario associated with the search request.

[0146] The units included in device 600 can be implemented in various ways, including software, hardware, firmware, or any combination thereof. In some embodiments, one or more units may be implemented using software and / or firmware, such as machine-executable instructions stored on a storage medium. In addition to or as an alternative to machine-executable instructions, some or all of the units in device 600 may be implemented at least partially by one or more hardware logic components. By way of example and not limitation, exemplary types of hardware logic components that may be used include field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), system-on-a-chip (SoCs), complex programmable logic devices (CPLDs), and so on.

[0147] Figure 7 shows a block diagram of an electronic device 700 in which one or more embodiments of the present disclosure may be implemented. It should be understood that the electronic device 700 shown in Figure 7 is merely exemplary and should not constitute any limitation on the functionality and scope of the embodiments described herein. The electronic device 700 shown in Figure 7 can be used to implement the electronic device 110 shown in Figure 1.

[0148] As shown in Figure 7, the electronic device 700 is in the form of a general-purpose electronic device. Components of the electronic device 700 may include, but are not limited to, one or more processors or processing units 710, memory 720, storage devices 730, one or more communication units 740, one or more input devices 750, and one or more output devices 760. The processing unit 710 may be a physical or virtual processor and is capable of performing various processes according to programs stored in the memory 720. In a multiprocessor system, multiple processing units execute computer-executable instructions in parallel to improve the parallel processing capability of the electronic device 700.

[0149] Electronic device 700 typically includes multiple computer storage media. Such media can be any accessible media that is accessible to electronic device 700, including but not limited to volatile and non-volatile media, removable and non-removable media. Memory 720 can be volatile memory (e.g., registers, cache, random access memory (RAM)), non-volatile memory (e.g., read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory), or some combination thereof. Storage device 730 can be a removable or non-removable medium and can include machine-readable media, such as flash drives, disks, or any other media that can be used to store information and / or data (e.g., training data for training) and can be accessed within electronic device 700.

[0150] Electronic device 700 may further include additional removable / non-removable, volatile / non-volatile storage media. Although not shown in FIG. 7, disk drives for reading from or writing to removable, non-volatile disks (e.g., "floppy disks") and optical disk drives for reading from or writing to removable, non-volatile optical disks may be provided. In these cases, each drive may be connected to a bus (not shown) via one or more data media interfaces. Memory 720 may include computer program product 725 having one or more program modules configured to perform various methods or actions of various embodiments of the present disclosure.

[0151] The communication unit 740 enables communication with other electronic devices via a communication medium. Additionally, the functionality of the components of the electronic device 700 can be implemented using a single computing cluster or multiple computing machines capable of communicating via communication connections. Therefore, the electronic device 700 can operate in a networked environment using logical connections to one or more other servers, network personal computers (PCs), or another network node.

[0152] Input device 750 can be one or more input devices, such as a mouse, keyboard, trackball, etc. Output device 760 can be one or more output devices, such as a monitor, speaker, printer, etc. Electronic device 700 can also communicate with one or more external devices (not shown) via communication unit 740 as needed. These external devices include storage devices, display devices, etc., and can communicate with one or more devices that enable user interaction with electronic device 700, or with any device that enables electronic device 700 to communicate with one or more other electronic devices (e.g., network card, modem, etc.). Such communication can be performed via input / output (I / O) interface (not shown).

[0153] According to an exemplary implementation of this disclosure, a computer-readable storage medium is provided that stores computer-executable instructions thereon, wherein the computer-executable instructions are executed by a processor to implement the methods described above. According to an exemplary implementation of this disclosure, a computer program product is also provided, which is tangibly stored on a non-transitory computer-readable medium and includes computer-executable instructions, which are executed by a processor to implement the methods described above.

[0154] Various aspects of this disclosure are described herein with reference to flowchart illustrations and / or block diagrams of methods, apparatuses, devices, and computer program products implemented according to this disclosure. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer-readable program instructions.

[0155] These computer-readable program instructions can be provided to a processing unit of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine such that, when executed by the processing unit of the computer or other programmable data processing apparatus, they create means for implementing the functions / actions specified in one or more blocks of the flowchart and / or block diagram. These computer-readable program instructions can also be stored in a computer-readable storage medium that causes a computer, programmable data processing apparatus, and / or other device to operate in a particular manner. Thus, the computer-readable medium storing the instructions comprises an article of manufacture that includes instructions for implementing aspects of the functions / actions specified in one or more blocks of the flowchart and / or block diagram.

[0156] Computer-readable program instructions can be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions that execute on the computer, other programmable data processing apparatus, or other device to perform the functions / actions specified in one or more boxes of a flowchart and / or block diagram.

[0157] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of an instruction, which contains one or more executable instructions for implementing the specified logical function. In some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0158] Various implementations of this disclosure have been described above. These descriptions are exemplary and not exhaustive, nor are they limited to the disclosed implementations. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described implementations. The terminology used herein is chosen to best explain the principles, practical applications, or improvements to technology in the market, or to enable others skilled in the art to understand the various implementations disclosed herein.

Claims

1. An information retrieval method, comprising: Based on the received search request, determine the target tag corresponding to the search request; Based on a comparison between the target tag and a set of tags associated with a set of points of interest, at least one point of interest matching the target tag is determined from the set of points of interest, wherein the set of tags is generated based at least on user interaction information associated with the set of points of interest; and Based on the at least one point of interest, determine the search results for the search request.

2. The method of claim 1, wherein the set of tags corresponding to the set of points of interest is determined based on the following process: Based on the user interaction information, descriptive text about the set of points of interest is generated; Based on the descriptive text and the metadata of the set of points of interest, determine the set of tags associated with the set of points of interest.

3. The method of claim 2, wherein generating descriptive text about the set of points of interest based on the user interaction information comprises: The user interaction information is provided to the first model to generate multiple data blocks; as well as The multiple data blocks are provided to the second model to generate the descriptive text about the set of points of interest.

4. The method of claim 2, wherein the descriptive text is further determined based on media content associated with the set of points of interest.

5. The method of claim 2, wherein determining the set of tags associated with the set of interests based on the descriptive text and the metadata of the set of interests comprises: The descriptive text and metadata of the set of interest points are provided to the third model to generate multiple candidate tags associated with the set of interest points; as well as Based on the multiple candidate tags, the set of tags associated with the set of points of interest is determined.

6. The method of claim 5, wherein determining the set of tags associated with the set of points of interest based on the plurality of candidate tags comprises: In response to the first evaluation information of the target candidate label satisfying the preset conditions, the target candidate label is added to the label set.

7. The method of claim 1, wherein determining at least one interest point matching the target tag from the set of interest points based on a comparison between the target tag and a set of tags associated with a set of interest points comprises: In response to the similarity between the target tag and the first tag in the tag set being greater than a first threshold but less than a second threshold, the target tag is replaced with the first tag; as well as Based on the comparison between the first tag and the tag set, the at least one point of interest is determined.

8. The method of claim 7, wherein the first label indicates at least one of the following: The name corresponding to the point of interest, The category corresponding to the point of interest The infrastructure corresponding to points of interest.

9. The method of claim 1, wherein the target tag includes a second tag and a third tag, and determining at least one interest point matching the target tag from the set of interest points based on a comparison of the target tag and a set of tags associated with a set of interest points comprises: Based on the comparison between the second tag and the tag set, a first group of candidate interest points is determined from the group of interest points; Based on the comparison between the third tag and the tag set, a second group of candidate interest points is determined from the first group of interest points; and Based on the first group of candidate interest points and the second group of candidate interest points, at least one interest point is determined.

10. The method of claim 9, wherein determining the at least one interest point based on the first group of candidate interest points and the second group of candidate interest points comprises: Based on the types of the second tag and the third tag, at least one interest point is determined from the first group of candidate interest points and the second group of candidate interest points.

11. The method of claim 1, wherein determining the search results for the search request based on the at least one point of interest comprises: Based on the at least one point of interest, determine a set of candidate points of interest that match the search request; as well as Based on the second evaluation information associated with the candidate interest point set, a set of target interest points is determined from the candidate interest point set as the search result.

12. The method of claim 11, wherein the candidate interest set further comprises a set of additional interest points determined based on at least one query term, the at least one query term being determined based on the search request.

13. The method of claim 11, wherein the second evaluation information associated with the candidate interest set is determined based on at least one of the following: The relevance of the search request to the target description information of the point of interest; The degree of matching between the search request and the points of interest; Business parameters associated with points of interest; The search scenario associated with the search request.

14. An apparatus for information retrieval, comprising: The first determining module is configured to determine the target tag corresponding to the received search request based on the received search request; as well as The second determining module is configured to determine at least one point of interest matching the target tag from the set of points of interest based on a comparison between the target tag and a set of tags associated with the set of points of interest, wherein the set of tags is generated based at least on user interaction information associated with the set of points of interest. as well as The third determining module is configured to determine the search results of the search request based on the at least one point of interest.

15. An electronic device comprising: At least one processing unit; as well as At least one memory, coupled to the at least one processing unit and storing instructions for execution by the at least one processing unit, which, when executed by the at least one processing unit, cause the electronic device to perform the method according to any one of claims 1 to 13.

16. A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method according to any one of claims 1 to 13.

17. A computer program product comprising computer-executable instructions, wherein the computer-executable instructions, when executed by a processor, implement the method according to any one of claims 1 to 13.

Images