Interest point query method and device, equipment and storage medium

Abstract

The invention discloses an interest point query method and device, equipment and a storage medium. The interest point query method comprises the steps of obtaining a to-be-queried text of a user; querying in a preset interest point cache library based on the to-be-queried text to obtain a candidate interest point list; determining an index score associated with each candidate interest point based on the behavior data associated with each candidate interest point in the candidate interest point list; and based on the index score associated with each candidate interest point, selecting a plurality of candidate interest points from the candidate interest point list and returning the candidate interest points to the user. According to the candidate interest point list queried in the interest point cache library according to the to-be-queried text, and the interest point cache library stores the query text and the interest point list associated with the query text, the time consumed by repeated retrieval can be reduced, and then the final candidate interest point is selected according to the behavior data associated with the candidate interest point, so that the search efficiency is improved. And the behavior data of the user is integrated into the retrieval recall process, so that the accuracy of interest point query is improved.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a point of interest query method, device, equipment and storage medium. Background Art

[0002] Currently, more and more users search for POIs (Point of Interest) in electronic maps, and the POI data stored in the database provides data support for POI searches.

[0003] The commonly used method for querying points of interest is to directly use the query word Query to search for POIs stored in the index library, and use the text matching function of Elasticsearch (ES) to query POIs when matching. The specific steps are as follows: Elasticsearch will perform text matching based on the Query and the POIs in the index library and calculate the relevance score. If the relevance scores of multiple POIs are the same, they are sorted according to the popularity of the POIs. However, each search requires full-text matching from the index library, which involves a large amount of data reading and calculation, especially when the amount of data is large, the response time becomes longer and the user experience decreases. In addition, the current solution mainly relies on the accuracy of text matching without considering the actual behavior data of users. It may recall some POIs that are related in text but not actually of interest to users, resulting in a decrease in the accuracy of search results. Summary of the invention

[0004] Based on this, it is necessary to provide a point of interest query method, device, equipment and storage medium for the above technical problems to solve at least one of the above technical problems.

[0005] The present invention provides a method for querying points of interest, comprising:

[0006] Get the user's query text;

[0007] Based on the text to be queried, querying in a preset interest point cache library to obtain a candidate interest point list, wherein the interest point cache library stores the query text, the interest point list associated with the query text, and behavior data associated with each interest point;

[0008] Determining an indicator score associated with each candidate point of interest based on the behavior data associated with each candidate point of interest in the list of candidate points of interest;

[0009] Based on the index score associated with each of the candidate points of interest, multiple candidate points of interest are selected from the candidate points of interest list and pushed to the user.

[0010] Optionally, according to a method for querying points of interest provided by the present invention, before querying in a preset points of interest cache based on the text to be queried to obtain a list of candidate points of interest, the method further includes:

[0011] Obtaining a user's search behavior log within a preset time period, wherein the search behavior log includes a plurality of query texts, a list of points of interest associated with the query texts, and behavior data associated with each point of interest;

[0012] For each of the query texts: expanding the query text to obtain a plurality of extended texts corresponding to the query text, and using the list of points of interest associated with the query text as the list of points of interest associated with each of the extended texts;

[0013] Determine the index score corresponding to each point of interest based on the behavior data corresponding to each point of interest associated with the extended text;

[0014] Each of the query texts, the list of interest points associated with the query text and a plurality of extended texts, and the index score associated with each interest point are stored in the interest point cache.

[0015] Optionally, according to a method for querying points of interest provided by the present invention, the behavior data includes exposure, clicks and transaction orders associated with the points of interest; the indicator scores include click-through rate and conversion rate;

[0016] The determining, based on the behavior data corresponding to each point of interest associated with each of the extended texts, an indicator score corresponding to each point of interest includes:

[0017] Aggregate the behavior data corresponding to each point of interest associated with the extended text to obtain the total exposure, total clicks and total transaction orders of each point of interest;

[0018] Based on the total exposure and total clicks associated with the points of interest, the click-through rate associated with the points of interest is determined; and based on the total clicks and total completed orders associated with the points of interest, the conversion rate associated with the points of interest is determined.

[0019] Optionally, according to a method for querying points of interest provided by the present invention, querying in a preset point of interest cache based on the text to be queried to obtain a list of candidate points of interest includes:

[0020] Expanding the text to be queried to obtain multiple expanded query texts;

[0021] Based on each of the extended query texts, a query is performed in the interest point cache to obtain a first query result corresponding to each of the extended texts;

[0022] Based on the first query results corresponding to each extended text, form the candidate point-of-interest list.

[0023] Optionally, according to a point-of-interest query method provided by the present invention, the forming of the candidate point-of-interest list based on the first query results corresponding to each extended text includes:

[0024] Perform text matching between the text to be queried and each point of interest in a preset database to obtain the relevance scores of each point of interest;

[0025] Based on the relevance scores, select multiple points of interest in the preset database as the second query results;

[0026] Perform redundant merging on the first query results and the second query results to obtain the candidate point-of-interest list.

[0027] Optionally, according to a point-of-interest query method provided by the present invention, before expanding the text to be queried to obtain multiple extended query texts, it further includes:

[0028] Perform error correction processing on the text to be queried;

[0029] Perform component recognition on the text to be queried after error correction processing, and based on the component recognition results, perform filtering processing on the text to be queried after error correction processing to obtain the filtered text to be queried.

[0030] Optionally, according to a point-of-interest query method provided by the present invention, the selecting of multiple candidate points of interest from the candidate point-of-interest list based on the index scores associated with each candidate point of interest and pushing them to the user includes:

[0031] Sort each candidate point of interest based on the index scores associated with each candidate point of interest;

[0032] According to the sorting result, select multiple candidate points of interest and push them to the user.

[0033] The present invention also provides a point-of-interest query device, including:

[0034] An acquisition module, configured to acquire the text to be queried of the user;

[0035] A query module, configured to query in a preset point-of-interest cache library based on the text to be queried to obtain a candidate point-of-interest list, where the point-of-interest cache library stores query texts, the point-of-interest lists associated with the query texts, and the behavior data associated with each point of interest;

[0036] A determination module, configured to determine an index score associated with each candidate point of interest based on the behavior data associated with each candidate point of interest in the candidate point of interest list;

[0037] A return module, configured to select multiple candidate points of interest from the candidate point of interest list based on the index scores associated with each candidate point of interest and push them to the user.

[0038] The present invention also provides a computer device, including a memory, a processor, and computer-readable instructions stored in the memory and executable on the processor. When the processor executes the computer-readable instructions, the above-mentioned point of interest query method is implemented.

[0039] The present invention also provides one or more readable storage media storing computer-readable instructions. When the computer-readable instructions are executed by a processor, the above-mentioned point of interest query method is implemented.

[0040] The above-mentioned point of interest query method, device, device and storage medium include: obtaining a text to be queried by a user; querying in a preset point of interest cache library based on the text to be queried to obtain a candidate point of interest list, where the point of interest cache library stores query texts, a list of points of interest associated with the query texts, and behavior data associated with each point of interest; determining an index score associated with each candidate point of interest based on the behavior data associated with each candidate point of interest in the candidate point of interest list; selecting multiple candidate points of interest from the candidate point of interest list based on the index scores associated with each candidate point of interest and pushing them to the user. In the present invention, the point of interest cache library stores query texts and a list of points of interest associated with the query texts. By matching the text to be queried with the query texts in the point of interest cache library to query the candidate point of interest list, the time consumption caused by repeated retrieval can be reduced. Furthermore, based on the behavior data associated with the candidate points of interest, the final candidate points of interest are selected, and the user's behavior data is incorporated into the retrieval and recall process, which is beneficial to improving the accuracy of point of interest query. Description of the Drawings

[0041] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments of the present invention will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0042] Figure 1 is a flowchart of a point of interest query method in an embodiment of the present invention;

[0043] Figure 2It is a schematic structural diagram of a point of interest query device in an embodiment of the present invention;

[0044] Figure 3 It is a schematic diagram of a computer device in an embodiment of the present invention. Detailed implementation manners

[0045] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0046] The terms used in one or more embodiments of the present invention are only for the purpose of describing specific embodiments, and are not intended to limit one or more embodiments of the present invention. The singular forms "a", "the", and "said" used in one or more embodiments of the present invention are also intended to include the plural forms, unless the context clearly indicates otherwise. It should also be understood that the term "and / or" used in one or more embodiments of the present invention refers to and includes any or all possible combinations of one or more related listed items.

[0047] In one embodiment, specifically, as Figure 1 shown, Figure 1 It is a schematic flowchart of a point of interest query method in an embodiment of the present invention. The embodiment of the present invention provides a point of interest query method, including the following steps:

[0048] Step S11, obtain the text to be queried of the user;

[0049] It should be noted that the text to be queried corresponding to different application scenarios is different. For example, in the e-commerce industry, the text to be queried may be the product name, and in the transportation field, the text to be queried is information such as the starting address input by the user.

[0050] Step S12, based on the text to be queried, query in a preset point of interest cache library to obtain a candidate point of interest list;

[0051] It should be noted that the point of interest cache library stores information such as several query texts, the list of points of interest associated with the query text, and the behavior data associated with each point of interest. The list of points of interest associated with the query text refers to the set of points of interest recommended by the system to the user after the user inputs the query text.

[0052] It should be noted that, in order to improve the accuracy of interest point query, the text to be queried is pre-corrected to ensure the accuracy of the text. Then, the corrected text to be queried is subjected to component recognition. For example, an NER model is used to recognize the components of the corrected text. Further, based on the component recognition results, the corrected text to be queried is filtered. For example, information such as provincial and municipal addresses in the text to be queried is removed to obtain the filtered text to be queried.

[0053] Specifically, in one embodiment, the filtered text to be queried is text-matched with each query text in the interest point cache library, and the list of interest points associated with the matched query text is used as the candidate interest point list.

[0054] In another embodiment, in order to improve the query accuracy, the text to be queried is first expanded to obtain multiple expanded query texts. For example, for the expansion of Hongyuan Shouzhu in Haidian District, Beijing: First, the provincial, municipal, and district address components are removed step by step to obtain the expanded query texts: Hongyuan Shouzhu in Haidian District and Hongyuan Shouzhu. Then, prefix generalization is further performed on Hongyuan Shouzhu in Haidian District and Hongyuan Shouzhu to obtain the expanded query texts: Hai, Haidian, Hong, Hongyuan, Shouzhu, etc. Therefore, the expanded query texts include Hai, Haidian, Hong, Hongyuan, Shouzhu, Hongyuan Shouzhu, Hongyuan Shouzhu in Haidian District, and Hongyuan Shouzhu in Haidian District, Beijing. Further, each of the expanded query texts is respectively text-matched with the query texts in the interest point cache library to obtain the first query result corresponding to each expanded text. Then, based on the first query result corresponding to each expanded text, the candidate interest point list is formed.

[0055] In other embodiments, the text to be queried can also be text-matched with each interest point in the preset database to obtain the correlation scores between the text to be queried and each of the interest points. Optionally, an index is created using the API of Elasticsearch to store the interest point data, and then the text matching is performed using the search API of Elasticsearch. Further, based on the correlation scores, multiple interest points are selected from the preset database as the second query result. Then, the first query result and the second query result are redundantly merged to obtain the candidate interest point list.

[0056] Step S13, based on the behavior data associated with each candidate interest point in the candidate interest point list, determine the index score associated with each candidate interest point;

[0057] It should be noted that the index score includes indicators such as click-through rate and conversion rate. Specifically, the click-through rate associated with the candidate point of interest is determined based on the exposure and click volume associated with the candidate point of interest, and the conversion rate associated with the candidate point of interest is determined based on the click volume and transaction volume associated with the candidate point of interest. In other embodiments, the point of interest cache library caches the index scores associated with each point of interest, and the index scores associated with the candidate point of interest are directly queried in the point of interest cache library.

[0058] Step S14: based on the index score associated with each of the candidate points of interest, multiple candidate points of interest are selected from the candidate points of interest list and pushed to the user.

[0059] Specifically, based on the index score associated with each of the candidate interest points, the candidate interest points are sorted, and further, based on the sorting result, multiple candidate interest points with higher index scores are selected, and the selected candidate interest points are pushed to the user.

[0060] The embodiment of the present invention, through the above scheme, includes: obtaining the user's text to be queried; based on the text to be queried, querying in a preset interest point cache library to obtain a list of candidate interest points, wherein the interest point cache library stores the query text, the interest point list associated with the query text, and the behavior data associated with each interest point; based on the behavior data associated with each candidate interest point in the candidate interest point list, determining the index score associated with each candidate interest point; based on the index score associated with each candidate interest point, selecting multiple candidate interest points in the candidate interest point list and pushing them to the user. The interest point cache library in the embodiment of the present invention stores the query text and the interest point list associated with the query text. By matching the text to be queried with the query text in the interest point cache library to query and obtain the candidate interest point list, the time consumption caused by repeated retrieval can be reduced, and then the final candidate interest point is selected according to the behavior data associated with the candidate interest point, and the user's behavior data is integrated into the retrieval and recall process, which is conducive to improving the accuracy of interest point query.

[0061] In one embodiment of the present invention, based on the text to be queried, before searching in a preset POI cache library to obtain a list of candidate POIs, the method further includes:

[0062] Step S21, obtaining a user's search behavior log within a preset time period, wherein the search behavior log includes a plurality of query texts, a list of points of interest associated with the query texts, and behavior data associated with each point of interest;

[0063] Step S21, for each query text: expanding the query text to obtain a plurality of extended texts corresponding to the query text, and using the list of points of interest associated with the query text as the list of points of interest associated with each extended text;

[0064] Step S22, determining an index score corresponding to each of the interest points associated with each of the extended texts based on the behavior data corresponding to the interest points;

[0065] Step S23: storing each of the query texts, the list of interest points associated with the query text and multiple extended texts, and the index score associated with each interest point in the interest point cache.

[0066] It should be noted that the behavior data includes information such as exposure, clicks, and transaction volume of the POI, among which exposure refers to the total number of times the POI appears in all search results and recommendation lists within a preset time period. Clicks refer to the total number of times users click on the POI within a preset time period. Transaction volume refers to the total number of times users complete transactions by clicking on the POI within a preset time period.

[0067] Specifically, the user's search behavior log within a preset time period is collected, and further, for each query text: the query text is expanded to obtain multiple extended texts corresponding to the query text, wherein the text expansion process is similar to the text expansion process of the above step S12, and will not be repeated here. In addition, the list of points of interest associated with the query text is used as the list of points of interest associated with each extended text; based on the behavior data corresponding to each point of interest associated with the extended text, the index score corresponding to each point of interest is calculated, for example, based on the total exposure and total clicks associated with the point of interest, the click-through rate associated with the point of interest is determined, and based on the total clicks and total transaction orders associated with the point of interest, the conversion rate associated with the point of interest is determined; further, each query text, the list of points of interest associated with the query text and multiple extended texts, and the index score associated with each point of interest are stored in the point of interest cache.

[0068] Through the above scheme, the embodiment of the present invention realizes determining the index score corresponding to each point of interest by combining the user's retrieval behavior log. During the retrieval and recall process, the index scores corresponding to the points of interest can be combined to determine the points of interest recommended to the user, thereby integrating the user's behavior data into the retrieval and recall process, which is conducive to improving the accuracy of interest point queries.

[0069] In one embodiment of the present invention, determining the index score corresponding to each point of interest based on the behavior data corresponding to each point of interest associated with the extended text includes:

[0070] The behavioral data corresponding to each point of interest associated with the extended text are aggregated to obtain the total exposure, total clicks and total transaction orders of each point of interest; based on the total exposure and total clicks associated with the point of interest, the click rate associated with the point of interest is determined; and based on the total clicks and total transaction orders associated with the point of interest, the conversion rate associated with the point of interest is determined.

[0071] Specifically, the exposure, clicks and transaction orders of each point of interest are aggregated. For example, referring to Table 1, the exposure, clicks and transaction orders of points of interest 1, 2 and 3 are aggregated to obtain the total exposure, clicks and transaction orders of each point of interest, among which, point of interest 1: total exposure: 1000+800=1800; total clicks: 200+150=350; total transaction orders: 50+40=90.

[0072] Point of interest 2: Total exposure: 1500+1800=3300; total clicks: 300+360=660; total transaction orders: 75+90=165.

[0073] Point of interest 3: Total exposure: 1200+1000=2200; Total clicks: 240+200=440; Total transaction orders: 60+50=110.

[0074] Extended text Point of interest Exposure Click-through rate Number of completed orders Text A Point of interest 1 1000 200 50 Text A Point of interest 2 1500 300 75 Text B Point of interest 1 800 150 40 Text B Point of interest 3 1200 240 60 Text C Point of interest 2 1800 360 90 Text C Point of interest 3 1000 200 50

[0075] Further, based on the total exposure and total clicks associated with the point of interest, the click rate associated with the point of interest is calculated, where click rate = total clicks / total exposure. In addition, based on the total clicks and total number of completed orders associated with the point of interest, the conversion rate associated with the point of interest is determined, where conversion rate = total number of completed orders / total number of clicks.

[0076] The embodiment of the present invention realizes, through the above scheme, aggregation processing of the behavior data corresponding to each point of interest associated with the extended text, and then calculation of indicators such as click-through rate and conversion rate associated with the point of interest based on the total exposure, total clicks and total transaction orders associated with the point of interest. Subsequently, the points of interest are recalled and sorted based on indicators such as click-through rate and conversion rate, thereby integrating the user's behavior data into the retrieval and recall process, effectively improving the accuracy of interest point queries.

[0077] In one embodiment of the present invention, based on the text to be queried, a query is performed in a preset interest point cache library to obtain a list of candidate interest points, including:

[0078] Step S31, expanding the text to be queried to obtain multiple expanded query texts;

[0079] Step S32: Query in the point of interest cache library based on each of the extended query texts to obtain a first query result corresponding to each extended text;

[0080] Step S33: Form the candidate point of interest list based on the first query result corresponding to each extended text.

[0081] Specifically, extend the text to be queried to obtain multiple extended query texts; further, perform text matching queries on each of the extended query texts and each query text in the point of interest cache library respectively to obtain a first query result corresponding to each extended text. Further, directly form the candidate point of interest list based on the first query result corresponding to each extended text.

[0082] In other embodiments, forming the candidate point of interest list based on the first query result corresponding to each extended text includes:

[0083] Perform text matching on the text to be queried and each point of interest in the preset database to obtain the relevance scores of each point of interest; based on the relevance scores, select multiple points of interest in the preset database as the second query result; perform redundant merging on the first query result and the second query result to obtain the candidate point of interest list.

[0084] Specifically, the text to be queried can also be matched with each point of interest in the preset database, and the relevance score between the text to be queried and each point of interest is calculated. Optionally, TF-IDF and cosine similarity are used to calculate the relevance score. Then, sort the relevance scores of each point of interest, and select the points of interest with higher scores as the second query result according to the sorting result of the relevance scores. Further, perform redundant merging on the first query result and the second query result to remove duplicate points of interest and obtain the candidate point of interest list.

[0085] Through the above solution, the embodiment of the present invention realizes that by extending the text to be queried, and then using each extended query text for query, a recall set relatively relevant to the text to be queried is obtained, effectively improving the accuracy of point of interest query.

[0086] It should be understood that the magnitudes of the sequence numbers of the steps in the above embodiments do not mean the order of execution. The execution order of each process should be determined according to its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present invention.

[0087] In one embodiment, a point of interest query device is provided, and the point of interest query device corresponds one-to-one with the point of interest query method in the above embodiment. As Figure 2 shownFigure 2 : is a schematic diagram of a structure of an interest point query device in an embodiment of the present invention, the interest point query device comprises:

[0088] An acquisition module 41 is used to acquire the text to be queried by the user;

[0089] A query module 42 is used to query a preset interest point cache based on the query text to obtain a candidate interest point list, wherein the interest point cache stores the query text, the interest point list associated with the query text, and the behavior data associated with each interest point;

[0090] A determination module 43, configured to determine an index score associated with each candidate interest point in the candidate interest point list based on the behavior data associated with each candidate interest point in the candidate interest point list;

[0091] The return module 44 is configured to select a plurality of candidate interest points from the candidate interest point list based on the index score associated with each of the candidate interest points and push the selected candidate interest points to the user.

[0092] The point of interest query device also includes:

[0093] Obtaining a user's search behavior log within a preset time period, wherein the search behavior log includes a plurality of query texts, a list of points of interest associated with the query texts, and behavior data associated with each point of interest;

[0094] For each of the query texts: expanding the query text to obtain a plurality of extended texts corresponding to the query text, and using the list of points of interest associated with the query text as the list of points of interest associated with each of the extended texts;

[0095] Determine the index score corresponding to each point of interest based on the behavior data corresponding to each point of interest associated with the extended text;

[0096] Each of the query texts, the list of interest points associated with the query text and a plurality of extended texts, and the index score associated with each interest point are stored in the interest point cache.

[0097] The point of interest query device also includes:

[0098] Aggregate the behavior data corresponding to each point of interest associated with the extended text to obtain the total exposure, total clicks and total transaction orders of each point of interest;

[0099] Based on the total exposure and total clicks associated with the points of interest, the click-through rate associated with the points of interest is determined; and based on the total clicks and total completed orders associated with the points of interest, the conversion rate associated with the points of interest is determined.

[0100] The query module 42 is further configured to:

[0101] Expand the text to be queried to obtain multiple expanded query texts;

[0102] Based on each of the expanded query texts, query in the point of interest cache library to obtain a first query result corresponding to each expanded text;

[0103] Based on the first query result corresponding to each expanded text, form the candidate point of interest list.

[0104] The query module 42 is further configured to:

[0105] Perform text matching between the text to be queried and each point of interest in the preset database to obtain a relevance score for each point of interest;

[0106] Based on the relevance score, select multiple points of interest in the preset database as the second query result;

[0107] Redundantly merge the first query result and the second query result to obtain the candidate point of interest list.

[0108] The point of interest query device further includes:

[0109] Perform error correction processing on the text to be queried;

[0110] Perform component recognition on the text to be queried after error correction processing, and based on the component recognition result, perform filtering processing on the text to be queried after error correction processing to obtain the text to be queried after filtering processing.

[0111] The return module 44 is further configured to:

[0112] Sort the candidate points of interest based on the index score associated with each candidate point of interest;

[0113] According to the sorting result, select multiple candidate points of interest and push them to the user.

[0114] For the specific limitations of the point of interest query device, reference can be made to the limitations of the point of interest query method in the above text, which will not be elaborated here. Each module in the above point of interest query device can be implemented in whole or in part by software, hardware, and their combination. The above modules can be embedded in or independent of the processor in the computer device in the form of hardware, or stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0115] In one embodiment, a computer device is provided. The computer device can be a server, and its internal structure diagram can be as Figure 3 shownFigure 3 It is a schematic diagram of a computer device in an embodiment of the present invention. The computer device includes a processor, a memory, a network interface, and a database connected through a device bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a readable storage medium and an internal memory. The readable storage medium stores an operating device, computer-readable instructions, and a database. The internal memory provides an environment for the operation of the operating device and computer-readable instructions in the readable storage medium. The database of the computer device is used to store data involved in the point of interest query method. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer-readable instructions are executed by the processor, a point of interest query method is implemented. The readable storage medium provided in this embodiment includes a non-volatile readable storage medium and a volatile readable storage medium.

[0116] In one embodiment, a computer device is provided. The computer device may be a terminal device, and its internal structure diagram may be as Figure 3 shown. The computer device includes a processor, a memory, and a network interface connected through a device bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a readable storage medium. The readable storage medium stores computer-readable instructions. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer-readable instructions are executed by the processor, a point of interest query method is implemented. The readable storage medium provided in this embodiment includes a non-volatile readable storage medium and a volatile readable storage medium.

[0117] In one embodiment, a computer device is provided, including a memory, a processor, and computer-readable instructions stored in the memory and executable on the processor. When the processor executes the computer-readable instructions, the steps of the point of interest query method as described above are implemented.

[0118] In one embodiment, a readable storage medium is provided. The readable storage medium stores computer-readable instructions. When the computer-readable instructions are executed by a processor, the steps of the above-mentioned point-of-interest query method are implemented. Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing relevant hardware through computer-readable instructions. The computer-readable instructions can be stored in a non-volatile readable storage medium or a volatile readable storage medium. When the computer-readable instructions are executed, they can include the processes of the embodiments of the above-mentioned various methods. Among them, any reference to a memory, storage, database, or other medium used in the various embodiments provided in the present application can include non-volatile and / or volatile memories. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or an external cache. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and Rambus dynamic RAM (RDRAM), etc.

[0119] Those skilled in the art can clearly understand that, for the convenience and simplicity of description, only the above-mentioned division of each functional unit and module is used as an example. In actual applications, the above-mentioned functions can be allocated to different functional units and modules according to needs, that is, the internal structure of the device is divided into different functional units or modules to complete all or part of the functions described above.

[0120] The above-mentioned embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the various embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. A method for querying points of interest, characterized in that: include: Get the user's query text; Based on the text to be queried, querying in a preset interest point cache library to obtain a list of candidate interest points, wherein the interest point cache library stores the query text, the interest point list associated with the query text, and behavior data associated with each interest point; Determining an indicator score associated with each candidate point of interest based on the behavior data associated with each candidate point of interest in the list of candidate points of interest; Based on the index score associated with each of the candidate points of interest, multiple candidate points of interest are selected from the candidate points of interest list and pushed to the user.

2. The method for querying points of interest according to claim 1, characterized in that: Before the querying in a preset POI cache based on the text to be queried to obtain a list of candidate POIs, the method further includes: Obtaining a user's search behavior log within a preset time period, wherein the search behavior log includes a plurality of query texts, a list of points of interest associated with the query texts, and behavior data associated with each point of interest; For each of the query texts: expanding the query text to obtain a plurality of extended texts corresponding to the query text, and using the list of points of interest associated with the query text as the list of points of interest associated with each of the extended texts; Determine the index score corresponding to each point of interest based on the behavior data corresponding to each point of interest associated with the extended text; Each of the query texts, the list of interest points associated with the query text and a plurality of extended texts, and the index score associated with each interest point are stored in the interest point cache.

3. The method for querying points of interest according to claim 2, characterized in that: The behavior data includes exposure, clicks and transaction orders associated with the point of interest; the indicator scores include click-through rate and conversion rate; The determining, based on the behavior data corresponding to each point of interest associated with each of the extended texts, an indicator score corresponding to each point of interest includes: Aggregate the behavior data corresponding to each point of interest associated with the extended text to obtain the total exposure, total clicks and total transaction orders of each point of interest; Based on the total exposure and total clicks associated with the points of interest, the click-through rate associated with the points of interest is determined; and based on the total clicks and total transaction orders associated with the points of interest, the conversion rate associated with the points of interest is determined.

4. The method for querying points of interest according to claim 1, characterized in that: The step of searching a preset POI cache based on the text to be queried to obtain a list of candidate POIs includes: Expanding the text to be queried to obtain multiple expanded query texts; Based on each of the extended query texts, a query is performed in the interest point cache to obtain a first query result corresponding to each of the extended texts; The candidate interest point list is formed based on the first query result corresponding to each extended text.

5. The method for querying points of interest according to claim 4, characterized in that: The forming the candidate interest point list based on the first query result corresponding to each extended text includes: Performing text matching between the query text and each interest point in a preset database to obtain a relevance score for each interest point; Based on the relevance score, selecting a plurality of points of interest in a preset database as a second query result; The first query result and the second query result are redundantly merged to obtain the candidate interest point list.

6. The method for querying points of interest according to claim 4, characterized in that: Before the text to be queried is expanded to obtain a plurality of expanded query texts, the method further includes: Performing error correction processing on the text to be queried; The text to be queried after error correction is subjected to component recognition, and based on the component recognition result, the text to be queried after error correction is subjected to filtering to obtain the text to be queried after filtering.

7. The method for querying points of interest according to claim 1, characterized in that: The selecting a plurality of candidate points of interest from the list of candidate points of interest based on the index score associated with each of the candidate points of interest and pushing the selected points of interest to the user comprises: sorting the candidate points of interest based on the index score associated with each of the candidate points of interest; According to the sorting result, a plurality of candidate points of interest are selected and pushed to the user.

8. A point of interest query device, characterized in that: include: The acquisition module is used to obtain the user's query text; A query module, configured to query a preset interest point cache based on the query text to obtain a candidate interest point list, wherein the interest point cache stores the query text, the interest point list associated with the query text, and behavior data associated with each interest point; A determination module, configured to determine an indicator score associated with each candidate point of interest based on the behavior data associated with each candidate point of interest in the candidate point of interest list; The returning module is used to select a plurality of candidate points of interest from the list of candidate points of interest based on the index score associated with each of the candidate points of interest and push the selected points of interest to the user.

9. A computer device comprising a memory, a processor, and computer-readable instructions stored in the memory and executed on the processor, characterized in that: When the processor executes the computer-readable instructions, the method for querying points of interest as described in any one of claims 1 to 7 is implemented.

10. A readable storage medium having computer readable instructions stored thereon, characterized in that: When the computer-readable instructions are executed by a processor, the method for querying points of interest as described in any one of claims 1 to 7 is implemented.

Images