Interest place pushing method, electronic device, and computer-readable storage medium

Abstract

The application relates to the technical field of artificial intelligence, in particular to an interest place pushing method, an electronic device and a computer readable storage medium. The interest place pushing method provided by the application needs to acquire user information of a target user first, then acquire current place information of a current place where the target user is located and candidate place information of a plurality of candidate places based on the user information, obtain a plurality of place correlation data based on each candidate place information and the current place information, filter the plurality of candidate place information based on the user information and the plurality of place correlation data, obtain interest place information, and finally push the interest place information to the target user. Since the interest place pushing method provided by the application filters the plurality of place correlation data based on the user information, the interest place information is obtained, so that the interest place pushed in the process of pushing the interest place is more matched with the actual needs of the user, and the accuracy of place recommendation is improved.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence technology, and in particular to a method for pushing interest-based locations, an electronic device, and a computer-readable storage medium. Background Technology

[0002] With the development of people's living standards, the popularity of smart terminals such as mobile phones and tablets is increasing, and interest-based location recommendations have emerged with the development of various mobile application programs (APPs). Interest-based location recommendations refer to the intelligent recommendation of places that a user might be interested in, based on analysis of time, location, or historical user data, with the user's authorization. For example, after detecting that a user has eaten dinner at a restaurant, the application may intelligently recommend nearby bars, cinemas, and other leisure venues for the user to choose from.

[0003] In related technologies, push notification methods are typically implemented based on users' search history within applications. Users are then pushed locations related to their search queries. However, application search history may not accurately reflect users' current actual needs, resulting in content that is often repetitive and redundant. Therefore, how to ensure that the recommended locations better match users' actual needs and improve the accuracy of location recommendations has become a major challenge that the industry urgently needs to address. Summary of the Invention

[0004] This application aims to address at least one of the technical problems existing in the prior art. To this end, this application proposes a method, electronic device, and storage medium for recommending locations based on interests, which enables the recommended locations to better match the user's actual needs, thereby improving the accuracy of location recommendations.

[0005] A method for recommending places of interest according to a first aspect of this application includes:

[0006] Obtain user information of the target user;

[0007] Based on the user information, obtain the current location information of the target user's current location and the candidate location information of multiple candidate locations;

[0008] Based on each candidate location information and the current location information, multiple location association data are obtained, and the location association data corresponds one-to-one with the candidate location information;

[0009] Based on the user information and the associated data of multiple locations, the information of multiple candidate locations is filtered to obtain information of locations of interest;

[0010] The information about the places of interest is pushed to the target user.

[0011] According to some embodiments of this application, the step of filtering multiple candidate location information based on the user information and multiple location association data to obtain location information of interest includes:

[0012] Based on the user information and the associated data of multiple locations, the filtering criteria data for the target user are determined;

[0013] Based on the filtering criteria data, multiple candidate location information are filtered to obtain location information of interest.

[0014] According to some embodiments of this application, the current location information includes first geographic information and first tag information of the current location, and each candidate location information includes second geographic information and second tag information of the candidate location. The step of obtaining multiple location association data based on each candidate location information and the current location information includes:

[0015] Based on the first geographic information and the second geographic information of each candidate location, the geographic location between each candidate location and the current location is compared to obtain multiple directional association data, and the directional association data corresponds one-to-one with the candidate locations;

[0016] Based on the first tag information and the second tag information of each candidate location, the category tags between each candidate location and the current location are compared to obtain multiple tag association data, and the tag association data corresponds one-to-one with the candidate location;

[0017] The location association data is obtained based on the location association data and the tag association data.

[0018] According to some embodiments of this application, the user information includes the historical behavior data of the target user, and the step of determining the filtering condition data of the target user based on the user information and multiple location-related data includes:

[0019] Based on the location association data and the historical behavior data for each of the aforementioned locations, behavioral association data between each of the candidate locations and the historical behavior data is obtained;

[0020] The location-related data for each location is parsed to obtain multiple location-related data and multiple tag-related data.

[0021] The filtering criteria data are determined based on the location association data, the tag association data, or the behavior association data.

[0022] According to some embodiments of this application, determining the filtering condition data based on the location association data, the tag association data, or the behavior association data includes:

[0023] The location association data, the label association data, and the behavior association data are input into a preset neural network model;

[0024] The location-related data, the label-related data, and the behavior-related data are weighted and configured using the neural network model to determine the filtering condition data.

[0025] According to some embodiments of this application, the step of determining the filtering condition data by weighting the location-related data, the label-related data, and the behavior-related data via the neural network model includes:

[0026] The dimensionality reduction processing is performed on the location association data, the label association data, and the behavior association data respectively to obtain the location feature vector corresponding to the location association data, the label feature vector corresponding to the label association data, and the behavior feature vector corresponding to the behavior association data.

[0027] The orientation feature vector, the label feature vector, and the behavior feature vector are coupled to obtain the interest feature vector;

[0028] The orientation feature vector, label feature vector, and behavior feature vector of the interest feature vector are weighted according to the weight configuration rule to obtain the updated interest feature vector, wherein the weight configuration rule is obtained based on the historical behavior data;

[0029] The updated interest feature vector is used as the filtering condition data.

[0030] According to some embodiments of this application, the step of filtering multiple candidate location information based on the filtering condition data to obtain location information of interest includes:

[0031] Based on the updated interest feature vector, the location probability is calculated to obtain the interest probability feature of each candidate location;

[0032] Based on the interest probability features, the interest location information is selected from multiple candidate location information.

[0033] According to some embodiments of this application, the user information includes first geographical information and historical behavior data. The step of obtaining current location information of the target user's current location and candidate location information of multiple candidate locations based on the user information includes:

[0034] Based on the first geographic information, the current location information is obtained;

[0035] Based on the current location information and the historical behavior data, information is filtered in a preset location information database to obtain the candidate location information of multiple candidate locations.

[0036] Secondly, embodiments of this application provide an electronic device, including: a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the interest location push method as described in any one of the embodiments of the first aspect of this application.

[0037] Thirdly, embodiments of this application provide a computer-readable storage medium storing a computer program, which is executed by a processor to implement the interest location push method as described in any one of the embodiments of the first aspect of this application.

[0038] The interest-based location push method, electronic device, and storage medium according to the embodiments of this application have at least the following beneficial effects:

[0039] The interest-based location recommendation method provided in this application requires first obtaining the target user's user information, then, based on the user information, obtaining the current location information of the target user and multiple candidate locations, further obtaining multiple location-related data based on each candidate location information and the current location information, with each location-related data corresponding one-to-one with the candidate location information, and then, based on the user information and the multiple location-related data, filtering the multiple candidate location information to obtain interest-based location information, and finally pushing the interest-based location information to the target user. Because this interest-based location recommendation method first obtains the current location information of the target user and multiple candidate locations based on user information, then obtains multiple location-related data using the current location information and multiple candidate locations as clues, and further filters based on user information to obtain interest-based location information, it can make the recommended interest-based locations more closely match the user's actual needs, improving the accuracy of location recommendations.

[0040] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

[0041] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the description of the embodiments taken in conjunction with the following drawings, in which:

[0042] Figure 1 A flowchart illustrating the interest-based location recommendation method provided in this application embodiment;

[0043] Figure 2 for Figure 1 Flowchart of step S102;

[0044] Figure 3 for Figure 1 Flowchart of step S103;

[0045] Figure 4 for Figure 1 Flowchart of step S104;

[0046] Figure 5 for Figure 4 Flowchart of step S401;

[0047] Figure 6 for Figure 5 Flowchart of step S503;

[0048] Figure 7 for Figure 6 Flowchart of step S602;

[0049] Figure 8 for Figure 4 Flowchart of step S402;

[0050] Figure 9 This is a schematic diagram of the hardware structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0051] The embodiments of this application are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain this application, and should not be construed as limiting this application.

[0052] In the description of this application, "several" means one or more, "more than" means two or more, "greater than," "less than," and "exceeding" are understood to exclude the stated number, while "above," "below," and "within" are understood to include the stated number. The use of "first" and "second" in the description is merely for distinguishing technical features and should not be construed as indicating or implying relative importance, or implicitly indicating the number of indicated technical features, or implicitly indicating the order of the indicated technical features.

[0053] In the description of this application, it should be understood that the orientation descriptions, such as up, down, left, right, front, and back, are based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this application.

[0054] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "illustrative embodiment," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0055] In the description of this application, it should be noted that, unless otherwise explicitly defined, terms such as "setting," "installation," and "connection" should be interpreted broadly. Those skilled in the art can reasonably determine the specific meaning of the above terms in this application based on the specific content of the technical solution. Furthermore, the identification of specific steps in the following text does not imply a limitation on the order of steps or execution logic. The execution order and logic between each step should be understood and inferred from the content described in the embodiments.

[0056] First, let's analyze some of the terms used in this application:

[0057] Artificial intelligence (AI) is a new branch of computer science that studies, develops, and applies theories, methods, technologies, and systems to simulate, extend, and expand human intelligence. It aims to understand the essence of intelligence and produce intelligent machines that can react in a way similar to human intelligence. Research in this field includes robotics, speech recognition, image recognition, natural language processing, and expert systems. AI can simulate the information processes of human consciousness and thought. It utilizes digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceiving the environment, acquiring knowledge, and using that knowledge to achieve optimal results. Fundamental AI technologies generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing, operating / interactive systems, and mechatronics. AI software technologies mainly include computer vision, robotics, biometrics, speech processing, natural language processing, and machine learning / deep learning.

[0058] Information extraction is a text processing technique that extracts factual information such as entities, relationships, and events from natural language text and outputs it as structured data. Information extraction is a technique for extracting specific information from text data. Text data is composed of specific units, such as sentences, paragraphs, and chapters. Text information is composed of smaller, specific units, such as characters, words, phrases, sentences, paragraphs, or combinations of these units. Extracting noun phrases, names of people, and place names from text data is an example of text information extraction. Of course, text information extraction techniques can extract information of various types.

[0059] The interest-based location recommendation method provided in this application relates to the field of artificial intelligence technology. This method can be applied to a terminal, a server, or software running on either a terminal or a server. In some embodiments, the terminal can be a smartphone, tablet, laptop, desktop computer, etc.; the server can be configured as an independent 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, CDN, and big data and artificial intelligence platforms; the software can be an application implementing the interest-based location recommendation method, but is not limited to the above forms.

[0060] This application can be used in a wide variety of general-purpose or special-purpose computer system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices. This application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform specific tasks or implement specific abstract data types. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0061] With the development of people's living standards, the popularity of smart terminals such as mobile phones and tablets is increasing, and interest-based location recommendations have emerged with the development of various mobile application programs (APPs). Interest-based location recommendations refer to the intelligent recommendation of places that a user might be interested in, based on analysis of time, location, or historical user data, with the user's authorization. For example, after detecting that a user has eaten dinner at a restaurant, the application may intelligently recommend nearby bars, cinemas, and other leisure venues for the user to choose from.

[0062] In related technologies, push notification methods are typically implemented based on users' search history within applications. Users are then pushed locations related to their search queries. However, application search history may not accurately reflect users' current actual needs, resulting in content that is often repetitive and redundant. Therefore, how to ensure that the recommended locations better match users' actual needs and improve the accuracy of location recommendations has become a major challenge that the industry urgently needs to address.

[0063] This application aims to address at least one of the technical problems existing in the prior art. To this end, this application proposes a method, electronic device, and storage medium for recommending locations based on interests, which enables the recommended locations to better match the user's actual needs, thereby improving the accuracy of location recommendations.

[0064] The following explanation is based on the accompanying drawings.

[0065] Reference Figure 1 A method for recommending places of interest according to a first aspect of this application includes:

[0066] Step S101: Obtain the user information of the target user;

[0067] Step S102: Based on user information, obtain the current location information of the target user and the candidate location information of multiple candidate locations;

[0068] Step S103: Based on the information of each candidate location and the current location information, multiple location association data are obtained, and the location association data corresponds one-to-one with the candidate location information;

[0069] Step S104: Based on user information and data associated with multiple locations, filter the information of multiple candidate locations to obtain information of locations of interest;

[0070] Step S105: Push the information about places of interest to the target users.

[0071] It should be understood that the interest-based location recommendation method described in steps S101 to S105 of this application requires first obtaining the target user's user information, then obtaining the current location information of the target user's current location and the candidate location information of multiple candidate locations based on the user information, further obtaining multiple location-related data based on each candidate location information and the current location information, with each location-related data corresponding one-to-one with the candidate location information, and further filtering the multiple candidate location information based on the user information and the multiple location-related data to obtain interest-based location information, and finally pushing the interest-based location information to the target user. Because the interest-based location recommendation method of this application first obtains the current location information of the target user's current location and the candidate location information of multiple candidate locations based on user information, then obtains multiple location-related data using the current location information and the multiple candidate information as clues, and further filters the data based on user information to obtain interest-based location information, it can make the recommended interest-based locations more closely match the user's actual needs during the interest-based location recommendation process, thus improving the accuracy of location recommendations.

[0072] In step S101 of some embodiments, user information refers to relevant information of the target user. This user information may include, but is not limited to, the target user's current location, historical behavioral data, search history on the application, and scanning records, among other information related to the target user. It should be understood that in some embodiments of this application, user information can be obtained in various ways. For example, if the user information refers to the target user's current location, it can be obtained through the location function of the target user's smart terminal, or by scanning a store's unique QR code. Alternatively, if the user information refers to the target user's historical behavioral data, it can be obtained from the historical data of the target user's smart terminal. It should be understood that the methods for obtaining user information may include, but are not limited to, the specific embodiments described above.

[0073] In step S102 of some embodiments of this application, it is necessary to first obtain the current location information of the target user's current location and the candidate location information of multiple candidate locations based on user information. It should be clarified that the current location information refers to relevant information about the target user's current location, such as the address, latitude and longitude, and relative position to landmarks, etc. Candidate locations refer to locations that the user may be interested in, and candidate location information refers to relevant information about the candidate locations, such as their address, latitude and longitude, and relative position to landmarks, etc. It should be understood that obtaining the current location information of the target user based on user information can be achieved in various ways. For example, when the target user scans the QR code of restaurant A, the current location information of the target user (i.e., restaurant A) can be determined based on the target user's scanning record; another example is obtaining the current location information of the target user based on the positioning function on the target user's smart terminal. The methods for obtaining the current location information of the target user based on user information can include, but are not limited to, the specific embodiments mentioned above. It should be noted that, typically after identifying the target user's current location, multiple candidate locations are then obtained. For example, if the target user's current location is a restaurant, then candidate locations could be leisure and entertainment venues around the restaurant (such as shopping malls or parks), or places the target user frequently visits after previous meals. After identifying the candidate locations, further candidate location information (such as relative location, whether it is currently open, and public reviews) is obtained for each candidate location. It should be understood that, based on user information, the methods for obtaining the target user's current location information and multiple candidate location information can include, but are not limited to, the specific embodiments described above.

[0074] It is important to emphasize that when processing data related to user identity or characteristics, such as user information, user behavior data, user historical data, and user location information, user permission or consent is always obtained first. Furthermore, the collection, use, and processing of this data comply with relevant national and regional laws, regulations, and standards. In addition, when this application embodiment needs to obtain sensitive personal information from users, separate permission or consent from the user is obtained through pop-ups or redirects to confirmation pages. Only after explicitly obtaining the user's separate permission or consent is the necessary user-related data for the normal operation of this application embodiment acquired.

[0075] Reference Figure 2 In some embodiments, user information includes first geographic information and historical behavior data, and step S102 may include, but is not limited to, steps S201 to S202.

[0076] Step S201: Based on the first geographic information, obtain the current location information;

[0077] Step S202: Based on the current location information and historical behavior data, information is filtered in the preset location information database to obtain candidate location information for multiple candidate locations.

[0078] In step S201 of some embodiments of this application, it is necessary to first obtain the current location information based on the first geographic information. It should be noted that the first geographic information refers to the location information of the target user's current location. It should be understood that the first geographic information in the user information can be obtained based on the target user's scanning records, the location function on the target user's smart terminal, or other methods. It should be pointed out that after obtaining the first geographic information, the target user's current location can be clearly identified. It is important to emphasize that the current location information is the relevant information about the target user's current location. After identifying the target user's current location, the current location information can be obtained through the network, a preset database, or other methods. For example, the target user's current location can be set as a keyword, and a network search engine can be used to obtain a series of current location information such as the address, latitude and longitude, and relative position to landmarks. Similarly, a preset database or other methods can also be used to obtain a series of current location information such as the address, latitude and longitude, and relative position to landmarks.

[0079] In step S202 of some embodiments of this application, after obtaining the current location information, further, based on the current location information and historical behavior data, information is filtered in a preset location information database to obtain candidate location information for multiple candidate locations. It should be noted that historical behavior data refers to data reflecting the target user's historical behavior. The types of historical behavior data are diverse and may include, but are not limited to, the target user's QR code scanning records, the target user's search records on the application, and the target user's habitual movement trajectory records, etc. The preset location information database refers to a pre-set database that stores various preset location information. It should be pointed out that preset location information refers to relevant information about preset locations, such as the address, latitude and longitude, and relative location to landmarks, etc., of the preset location. In some embodiments of this application, all or part of the locations in the target user's area can be determined as preset locations, and the preset location information corresponding to each preset location can be queried from the network, and the preset location information can be further entered into the preset location information database.

[0080] In some specific embodiments of this application, if the target user is currently at restaurant A, the embodiments first need to obtain the name and address of restaurant A (i.e., current location information) based on its geographical location (first geographical information). Then, based on the location name and historical behavior data, information is filtered in a preset location information database to find m candidate locations the target user has visited after dining at restaurant A, and the operating hours and remaining parking spaces corresponding to these m candidate locations are obtained respectively (i.e., candidate location information). Alternatively, based on the location address, information is filtered in the preset location information database to find n candidate locations closest to the target user, and the public satisfaction ratings and promotional activities corresponding to these n candidate locations are obtained respectively (i.e., candidate location information). It should be understood that there are various ways to obtain the current location information of the target user's current location and the candidate location information of multiple candidate locations based on user information, including, but not limited to, the specific embodiments described above.

[0081] Through the above steps S201 to S202, based on the current location information and historical behavior data, multiple candidate locations that are more compatible with the target user's interests can be selected from the numerous preset location information included in the preset location information database, thereby improving the efficiency of determining the location information of interest from multiple candidate locations in subsequent steps.

[0082] In step S103 of some embodiments of this application, after obtaining the current location information of the target user's current location and the candidate location information of multiple candidate locations, further, based on each candidate location information and the current location information, multiple location-related data are obtained, and the location-related data corresponds one-to-one with the candidate location information. It should be noted that since the current location information is the relevant information of the target user's current location, such as the address, latitude and longitude, and relative position to landmarks of the target user's current location, and the candidate location information is the relevant information of the candidate locations, such as the address, latitude and longitude, and relative position to landmarks of the candidate locations, after obtaining the current location information and the candidate location information of multiple candidate locations, the current location information can be used as a benchmark to compare with the candidate location information of each candidate location, thereby obtaining some location-related data. It should be clarified that the location-related data corresponds one-to-one with the candidate location information, specifically meaning that each time the current location information is compared with the candidate location information of a candidate location, a location-related data is generated, where the location-related data refers to data reflecting the association information between the current location and the candidate locations. It should be understood that the types of location-related data are diverse, and may include, but are not limited to: the geographical distance between the current location and the target location, the public transportation routes between the current location and the target location, whether the target user frequently visits the target location when in the current location, whether the general public frequently visits the target location when in the current location, and a series of other data reflecting the association information between the current location and the candidate locations.

[0083] In some specific embodiments of this application, if the target user's current location is Restaurant A, then after obtaining the current location information of Restaurant A, it can be compared with the candidate location information corresponding to candidate locations B, C, and D respectively. Specifically, comparing the current location information with the candidate location data of candidate location B forms first location association data; comparing the current location information with the candidate location data of candidate location C forms second location association data; and comparing the current location information with the candidate location data of candidate location D forms third location association data. The first, second, and third location association data reflect the connections between Restaurant A and candidate locations B, C, and D respectively. For example, the first location association data might indicate that the distance between Restaurant A and candidate location B is 300 meters; the second location association data might indicate that the target user has visited candidate location C multiple times after dining at Restaurant A; and the third location association data might indicate that the user has repeatedly chosen candidate location D as their next destination after dining at Restaurant A.

[0084] Reference Figure 3In some embodiments, the current location information includes the first geographic information and the first tag information of the current location, and each candidate location information includes the second geographic information and the second tag information of a candidate location. Step S103 may include, but is not limited to, steps S301 to S303.

[0085] Step S301: Based on the first geographic information and the second geographic information of each candidate location, compare the geographic location between each candidate location and the current location to obtain multiple directional association data, and the directional association data corresponds one-to-one with the candidate locations;

[0086] Step S302: Based on the first label information and the second label information of each candidate location, compare the category label between each candidate location and the current location to obtain multiple label association data, and the label association data corresponds one-to-one with the candidate locations;

[0087] Step S303: Obtain location association data based on the orientation association data and the tag association data.

[0088] According to step S301 in some embodiments of this application, based on the first geographic information and the second geographic information of each candidate location, the geographical location between each candidate location and the current location is compared to obtain multiple location association data, and the location association data corresponds one-to-one with the candidate locations. It should be noted that the second geographic information refers to the location information of the candidate locations. Since the first geographic information refers to the location information of the target user's current location, based on the first geographic information and the second geographic information of each candidate location, the geographical location between each candidate location and the current location can be compared to obtain multiple location association data. It should be clarified that the location association data corresponds one-to-one with the candidate locations; specifically, each time the first geographic information is compared with a second geographic information, a location association data is generated, where the location association data refers to data reflecting the geographical location and direction between the current location and the candidate locations.

[0089] According to step S302 in some embodiments of this application, based on the first label information and the second label information of each candidate location, the category label between each candidate location and the current location is compared to obtain multiple label association data, and the label association data corresponds one-to-one with the candidate locations. It should be noted that the first label information refers to the label information of the target user's current location, and the second label information refers to the label information of the candidate locations. It should be understood that the label information can be a pre-defined classification of various locations in the target user's area. For example, restaurants, eateries, and food streets can be labeled with the label "catering," guesthouses, hotels, and inns can be labeled with the label "accommodation," and parks and lakeside trails can be labeled with the label "leisure." It should be pointed out that in some embodiments of this application, all or some locations in the target user's area are configured with corresponding labels. Therefore, based on the first label information and the second label information of each candidate location, the category label between each candidate location and the current location can be compared to obtain multiple label association data. It needs to be clarified that the tag association data corresponds one-to-one with the candidate locations. Specifically, each time the first tag information is compared with a second tag information, a tag association data is generated. The tag association data refers to the data that reflects the category tag association information between the current location and the candidate locations.

[0090] In step S303 of some embodiments of this application, location association data is obtained based on the orientation association data and the tag association data. It should be noted that the location association data in some embodiments of this application has several components, which may include, but are not limited to, orientation association data and tag association data.

[0091] Through the above steps S301 to S303, it should be understood that since the location-related data includes location-related data and tag-related data, the location-related data can at least be used to filter multiple candidate location information in subsequent steps based on two dimensions: "geographical distance" and "category tag" to obtain location information of interest. Therefore, in the process of pushing location of interest, the pushed location of interest can be further matched with the user's actual needs, thereby improving the accuracy of location recommendation.

[0092] Step S104: Based on user information and data associated with multiple locations, filter the information of multiple candidate locations to obtain information of locations of interest;

[0093] According to step S104 in some embodiments of this application, after obtaining multiple location association data, multiple candidate location information is filtered based on user information and the multiple location association data to obtain interest location information. It should be emphasized that user information refers to relevant information of the target user, which may include, but is not limited to: the target user's current location, the target user's historical behavior data, the target user's search history on the application, the target user's QR code scanning history, and a series of other information related to the target user. Location association data refers to data reflecting the association information between the current location and candidate locations. The types of location association data are diverse and may include, but are not limited to: the geographical distance between the current location and the target location, public transportation routes between the current location and the target location, whether the target user frequently visits the target location when in the current location, and whether the general public frequently visits the target location when in the current location, and a series of other data reflecting the association information between the current location and candidate locations. Therefore, based on user information and multiple location association data, the target user can be used as a benchmark. By combining the location association data between the target user's current location and multiple candidate locations, the candidate location information can be further filtered, ultimately obtaining interest location information that better matches the target user's interests, thus improving the accuracy of location recommendations.

[0094] Reference Figure 4 In some embodiments, step S104 may include, but is not limited to, steps S401 to S402.

[0095] Step S401: Determine the filtering criteria data for the target user based on user information and data related to multiple locations;

[0096] Step S402: Based on the filtering criteria data, filter the information of multiple candidate locations to obtain the location of interest information.

[0097] According to steps S401 to S402 in some embodiments of this application, the filtering criteria data for the target user are first determined based on user information and multiple location association data. Then, based on the filtering criteria data, multiple candidate location information is filtered to obtain interest location information. It should be noted that the filtering criteria data refers to the basis for filtering multiple candidate location information. The filtering criteria data needs to be further obtained based on user information and multiple location association data. This is because user information can reflect the needs of the target user, while location association data can reflect the relationship between the user's current location and multiple candidate locations. Therefore, by filtering candidate location information that is closer to the needs of the target user from multiple candidate location information based on user information, interest location information that better matches the user's interests can be obtained. Thus, in the process of pushing interest location information, the pushed interest location information can be further matched with the user's actual needs, improving the accuracy of location recommendation.

[0098] Reference Figure 5 In some embodiments, the user information includes the target user's historical behavior data, and step S401 may include, but is not limited to, steps S501 to S503.

[0099] Step S501: Based on the location association data and historical behavior data, obtain the behavior association data between each candidate location and the historical behavior data;

[0100] Step S502: parse the associated data for each location to obtain multiple location-related data and multiple tag-related data;

[0101] Step S503: Determine the filtering criteria data based on location-related data, tag-related data, or behavior-related data.

[0102] According to step S501 in some embodiments of this application, behavioral association data between each candidate location and historical behavioral data is obtained based on each location association data and historical behavioral data. It should be emphasized that location association data refers to data reflecting the association information between the current location and candidate locations. The types of location association data are diverse, including but not limited to: the geographical distance between the current location and the target location, public transportation routes between the current location and the target location, whether the target user frequently visits the target location when in the current location, and whether the general public frequently visits the target location when in the current location, etc. Furthermore, historical behavioral data refers to data reflecting the target user's historical behavior. The types of historical behavioral data are diverse, including but not limited to: the target user's QR code scanning records, the target user's search records on the application, and the target user's habitual movement trajectory records, etc. Clearly, the target user's historical behavioral data can reflect the target user's interests. Therefore, based on historical behavioral data, from the dimension of "user preferences," a reference can be provided for determining the filtering criteria data in subsequent steps, facilitating the filtering of multiple candidate location information to obtain information on locations of interest. It should be noted that behavioral correlation data refers to data obtained by comparing various candidate locations based on historical behavioral data. It should be understood that behavioral correlation data reflects the target user's preference for each candidate location. According to some embodiments provided in this application, the formation of behavioral correlation data can depend on various factors. For example, locations frequently visited by the target user, locations the target user would go to even if it's a longer route, and locations the target user would visit even in bad weather often have a higher degree of preference. Locations the target user never visits, and locations the target user has visited once and will never return to, often have a lower degree of preference. The degree of preference for different candidate locations is reflected in the behavioral correlation data. It should be clarified that there are various ways to obtain behavioral correlation data between each candidate location and historical behavioral data based on each location correlation data and historical behavioral data, including, but not limited to, the specific embodiments mentioned above.

[0103] According to step S502 in some embodiments of this application, each location-related data is parsed to obtain multiple location-related data and multiple tag-related data. It should be noted that if the first geographic information refers to the location information of the target user's current location and the second geographic information refers to the location information of candidate locations, then based on the first geographic information and the second geographic information of each candidate location, multiple location-related data can be obtained by comparing the geographical locations of each candidate location with the current location. It should be clarified that the location-related data corresponds one-to-one with the candidate locations. Specifically, each comparison of the first geographic information with a second geographic information generates one location-related data point, where the location-related data reflects the geographical location and direction between the current location and the candidate locations. Furthermore, the first tag information refers to the tag information of the target user's current location, and the second tag information refers to the tag information of the candidate locations. It should be understood that the tag information can be a pre-defined classification of various locations in the target user's area. For example, restaurants, eateries, and food streets can be tagged with "catering," guesthouses, hotels, and inns can be tagged with "accommodation," and parks and lakeside trails can be tagged with "leisure." It should be noted that in some embodiments of this application, all or some locations in the target user's area are configured with corresponding tags. Therefore, based on the first tag information and the second tag information of each candidate location, the category tags of each candidate location can be compared with the current location to obtain multiple tag association data. It should be clarified that the tag association data corresponds one-to-one with the candidate locations. Specifically, each time the first tag information is compared with a second tag information, a tag association data is generated, where the tag association data refers to the data reflecting the category tag association information between the current location and the candidate locations.

[0104] According to step S503 in some embodiments of this application, filtering condition data is determined based on location-related data, tag-related data, or behavior-related data. It should be noted that location-related data, tag-related data, or behavior-related data are all used to provide a reference for determining the filtering condition data.

[0105] It should be understood that, through the above steps S501 to S503, since the location-related data includes location-related data and tag-related data, the location-related data can at least provide a reference for determining the filtering conditions in subsequent steps based on two dimensions: "geographical distance" and "category tag". The target user's historical behavior data can reflect the target user's interests. Therefore, based on the historical behavior data, the "user preference" dimension can be used to further provide a reference for determining the filtering conditions in subsequent steps, which is convenient for filtering multiple candidate location information to obtain interest location information. Thus, in the process of pushing interest locations, the pushed interest locations can be more matched with the user's actual needs, improving the accuracy of location recommendations.

[0106] Reference Figure 6 In some embodiments, step S503 may include, but is not limited to, steps S601 to S602.

[0107] Step S601: Input the orientation association data, label association data, and behavior association data into the preset neural network model;

[0108] Step S602: The neural network model is used to configure the weights of the location-related data, label-related data, and behavior-related data to determine the filtering criteria data.

[0109] According to steps S601 to S602 in some embodiments provided in this application, after obtaining the location-related data, tag-related data, and behavior-related data, the location-related data, tag-related data, and behavior-related data can be further input into a preset neural network model. The neural network model then assigns weights to the location-related data, tag-related data, and behavior-related data to determine the filtering condition data. It should be noted that since interest-based location push refers to intelligently pushing locations that a user might be interested in based on time factors, location factors, or historical user data, with the user's authorization, in order to make interest-based location push more efficient and accurate, in some preferred embodiments of this application, a preset neural network model is used to assign weights to the location-related data, tag-related data, and behavior-related data, and determine the filtering condition data. The preset neural network model in some embodiments of this application may include, but is not limited to: deep learning models based on Convolutional Neural Networks (CNN), models based on Long Short-Term Memory (LSTM) neural networks combined with attention mechanisms, and BERT (Bidirectional Encoder Representation from Transformers) language models. It should be noted that the preset neural network model can be configured with reference to the target user's historical behavior data.

[0110] Reference Figure 7 In some embodiments, step S602 may include, but is not limited to, steps S701 to S704.

[0111] Step S701: Dimensionality reduction is performed on the directional association data, label association data, and behavior association data respectively to obtain the directional feature vector corresponding to the directional association data, the label feature vector corresponding to the label association data, and the behavior feature vector corresponding to the behavior association data.

[0112] Step S702: Couple the orientation feature vector, label feature vector, and behavior feature vector to obtain the interest feature vector;

[0113] Step S703: Adjust the weights of the orientation feature vector, label feature vector, and behavior feature vector of the interest feature vector based on the weight configuration rule to obtain the updated interest feature vector. The weight configuration rule is obtained based on historical behavior data.

[0114] Step S704: Use the updated interest feature vector as the filtering condition data.

[0115] According to step S701 in some embodiments of this application, dimensionality reduction processing is performed on the directional association data, label association data, and behavior association data respectively to obtain the directional feature vector corresponding to the directional association data, the label feature vector corresponding to the label association data, and the behavior feature vector corresponding to the behavior association data. It should be noted that the purpose of performing dimensionality reduction processing on the directional association data, label association data, and behavior association data respectively to obtain the directional feature vector corresponding to the directional association data, the label feature vector corresponding to the label association data, and the behavior feature vector corresponding to the behavior association data is to facilitate data processing by the neural network model, thereby improving the efficiency of weight configuration in the neural network model.

[0116] According to steps S702 to S704 in some embodiments of this application, the azimuth feature vector, label feature vector, and behavior feature vector are coupled to obtain an interest feature vector. Then, based on a weight configuration rule, the weights of the azimuth feature vector, label feature vector, and behavior feature vector of the interest feature vector are adjusted to obtain an updated interest feature vector. The weight configuration rule is obtained based on historical behavior data. Finally, the updated interest feature vector is used as the filtering condition data. It should be noted that after obtaining the azimuth feature vector, label feature vector, and behavior feature vector, it is necessary to configure the weights of the azimuth feature vector, label feature vector, and behavior feature vector. Before the weight configuration, the azimuth feature vector, label feature vector, and behavior feature vector need to be coupled and incorporated into the same feature vector, i.e., the interest feature vector. Then, the weights of the azimuth feature vector, label feature vector, and behavior feature vector of the interest feature vector are further adjusted based on the weight configuration rule to obtain an updated interest feature vector, so that the updated interest feature vector can be used as the filtering condition data. It needs to be clarified that the interest feature vector refers to the feature vector used to predict the interest orientation of the target user. The weight ratio of the location feature vector, the label feature vector, and the behavior feature vector is the prediction weight ratio of the target user's interest orientation.

[0117] In some exemplary embodiments of this application, the weight configuration rule based on historical behavior data refers to the following: after each push of interest location information, this embodiment further detects the target user's next target action. If the target user's next target action corresponds to the previously pushed interest location information, positive feedback information is obtained; otherwise, if the target user's next target action does not correspond to the previously pushed interest location information, negative feedback information is obtained. The positive and negative feedback information corresponding to several pushes is recorded and sent to the neural network model, so that the neural network model uses the recorded positive and negative feedback information as the basis for adjusting the weight configuration rule. As the number of pushes increases, the weight configuration rule of the neural network model can be continuously optimized. Therefore, the push of interest location information in this embodiment will increasingly conform to the user's interest orientation. It should be noted that in some preferred embodiments, if sufficient historical behavioral data is not yet available, the weight configuration rules of the neural network model will assign higher weights to the location feature vector and the label feature vector, emphasizing the correlation between the target user's location and candidate locations, so that the interest location information recommended by this method is more meaningful. As historical behavioral data accumulates, the weight ratio of the behavioral feature vector in the interest feature vector will increase, so that the filtering criteria data are closer to the target user's interests. It should be understood that there are various ways to configure the weight configuration rules, which may include, but are not limited to, the specific embodiments mentioned above.

[0118] Reference Figure 8 In some embodiments, step S402 may include, but is not limited to, steps S801 to S802.

[0119] Step S801: Calculate the location probability based on the updated interest feature vector to obtain the interest probability feature of each candidate location;

[0120] Step S802: Based on the interest probability features, select the interest location information from multiple candidate location information.

[0121] According to steps S801 to S802 in some embodiments of this application, after using the updated interest feature vector as the filtering condition data, the location probability can be further calculated based on the updated interest feature vector to obtain the interest probability feature of each candidate location. It should be understood that since the interest feature vector refers to the feature vector used to predict the target user's interest orientation, the weight ratio of the orientation feature vector, label feature vector, and behavior feature vector is the prediction weight ratio for the target user's interest orientation. Therefore, by calculating the location probability based on the updated interest feature vector, the interest probability feature of each candidate location can be obtained. In some embodiments of this application, the location probability calculation can be performed by mapping the prediction weight ratio in the interest feature vector to the interest probability feature as the output result using the Softmax function. It should be understood that different interest probability features will be obtained for different candidate locations. It should be noted that after obtaining the interest probability feature, interest location information can be filtered from multiple candidate location information using various methods. For example, each candidate location can be arranged according to the interest probability feature, and then the several candidate locations that best match the predicted target user's interest orientation can be determined as the interest location information. It should be understood that since interest feature vectors refer to feature vectors used to predict the interest orientation of target users, filtering interest location information from multiple candidate location information using interest feature vectors can further improve the accuracy of location recommendations by making the recommended interest locations more closely match the actual needs of users during the process of pushing interest locations.

[0122] In some specific embodiments of this application, a model based on a Long Short-Term Memory (LSTM) neural network combined with an attention mechanism is used as the neural network model in the embodiments of this application. It should be noted that the model based on Long Short-Term Memory (LSTM) neural network combined with an attention mechanism includes an embedding layer, an attention layer, a LSTM long short-term attention layer, and a classification output layer. The embedding layer performs dimensionality reduction on the location-related data, label-related data, and behavior-related data respectively, obtaining location feature vectors for location-related data, label feature vectors for label-related data, and behavior feature vectors for behavior-related data. These feature vectors are then coupled to obtain the interest feature vector. The attention layer adjusts the weights of the location, label, and behavior feature vectors based on weight configuration rules, resulting in an updated interest feature vector. The LSTM long short-term attention layer obtains positive and negative feedback information corresponding to the interest location information to update the weight configuration rules. The classification output layer calculates the location probability based on the updated interest feature vector, obtaining the interest probability feature for each candidate location, and then selects the interest location information from multiple candidate location information based on the interest probability feature.

[0123] In step S105 of some embodiments, pushing the location information of interest to the target user can be achieved in various ways, such as pushing the location information of interest to the user in the form of a pop-up window, pushing the location information of interest to the user in the message notification bar, and many other push methods. It should be emphasized that the location information of interest refers to candidate location information that matches the predicted interest orientation of the target user. It should be understood that the location information of interest can be the candidate location information that best matches the predicted interest orientation of the target user, or it can be multiple candidate location information that are relatively consistent with the predicted interest orientation of the target user. It should be clarified that the methods of pushing location information of interest to the target user can include, but are not limited to, the specific embodiments mentioned above.

[0124] Figure 9 An electronic device 900 according to an embodiment of this application is shown. The electronic device 900 includes a processor 901, a memory 902, and a computer program stored in the memory 902 and executable on the processor 901. When the computer program is executed, it is used to perform the above-described location of interest push method.

[0125] The processor 901 and memory 902 can be connected via a bus or other means.

[0126] The memory 902, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs, such as the interest-based location push method described in the embodiments of this application. The processor 901 implements the above-described interest-based location push method by running the non-transitory software program and instructions stored in the memory 902.

[0127] The memory 902 may include a program storage area and a data storage area, wherein the program storage area may store the operating system and applications required for at least one function. The data storage area may store the above-described location-of-interest push method. Furthermore, the memory 902 may include high-speed random access memory 902, and may also include non-transitory memory 902, such as at least one storage device, flash memory, or other non-transitory solid-state storage device. In some embodiments, the memory 902 may optionally include remotely located memories 902 relative to the processor 901, which can be connected to the electronic device 900 via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0128] The non-transient software program and instructions required to implement the above-described location-of-interest push method are stored in memory 902. When executed by one or more processors 901, the above-described location-of-interest push method is executed, for example, executing... Figure 1 Method steps S101 to S105, Figure 2 Method steps S201 to S202, Figure 3 Method steps S301 to S303, Figure 4 Method steps S401 to S402, Figure 5 Method steps S501 to S503, Figure 6 Method steps S601 to S602, Figure 7 Method steps S701 to S704 in the above method Figure 8 The method steps S801 to S802.

[0129] This application also provides a computer-readable storage medium storing computer-executable instructions for performing the above-described location of interest push method.

[0130] In one embodiment, the computer-readable storage medium stores computer-executable instructions that are executed by one or more control processors, for example, executing... Figure 1 Method steps S101 to S105, Figure 2 Method steps S201 to S202, Figure 3Method steps S301 to S303, Figure 4 Method steps S401 to S402, Figure 5 Method steps S501 to S503, Figure 6 Method steps S601 to S602, Figure 7 Method steps S701 to S704 in the above method Figure 8 The method steps S801 to S802.

[0131] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0132] Those skilled in the art will understand that all or some of the steps and systems in the methods disclosed above can be implemented as software, firmware, hardware, and suitable combinations thereof. Some or all of the physical components can be implemented as software executed by a processor, such as a central processing unit, digital signal processor, or microprocessor, or as hardware, or as an integrated circuit, such as an application-specific integrated circuit. Such software can be distributed on a computer-readable medium, which can include computer storage media (or non-transitory media) and communication media (or transient media). As is known to those skilled in the art, the term computer storage media includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules, or other data). Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technologies, CD-ROM, digital versatile disc (DVD) or other optical disc storage, magnetic cartridges, magnetic tape, storage device storage or other magnetic storage devices, or any other medium that can be used to store desired information and is accessible to a computer. Furthermore, as is known to those skilled in the art, communication media typically include computer-readable instructions, data structures, program modules, or other data in modulated data signals such as carrier waves or other transmission mechanisms, and may include any information delivery medium. It should also be understood that the various implementation methods provided in this application can be arbitrarily combined to achieve different technical effects.

[0133] The above provides a detailed description of the preferred embodiments of this application. However, this application is not limited to the above-described embodiments. Those skilled in the art can make various equivalent modifications or substitutions without departing from the spirit of this application. All such equivalent modifications or substitutions are included within the scope defined by the claims of this application.

Claims

1. A method for recommending locations based on interests, characterized in that, include: Obtain user information of the target user; wherein, the user information includes the target user's historical behavior data; Based on the user information, obtain the current location information of the target user's current location and the candidate location information of multiple candidate locations; Based on each candidate location information and the current location information, multiple location association data are obtained, and the location association data corresponds one-to-one with the candidate location information; Based on the location association data and the historical behavior data for each of the aforementioned locations, behavioral association data between each of the candidate locations and the historical behavior data is obtained; wherein, the behavioral association data is used to reflect the target user's preference for each candidate location; Each location-related data is parsed to obtain multiple location-related data and multiple tag-related data; wherein, the location-related data refers to data reflecting the geographical location and direction between the current location and the candidate location, and the tag-related data is used to reflect the category tag-related information between the current location and the candidate location; The location association data, the label association data, and the behavior association data are input into a preset neural network model; The dimensionality reduction of the location-related data, the label-related data, and the behavior-related data is performed by the neural network model to obtain the location feature vector corresponding to the location-related data, the label feature vector corresponding to the label-related data, and the behavior feature vector corresponding to the behavior-related data. The orientation feature vector, the label feature vector, and the behavior feature vector are coupled to obtain the interest feature vector; The orientation feature vector, label feature vector, and behavior feature vector of the interest feature vector are weighted according to the weight configuration rule to obtain the updated interest feature vector, wherein the weight configuration rule is obtained based on the historical behavior data; The updated interest feature vectors are used as filtering criteria data; Based on the filtering criteria data, multiple candidate location information are filtered to obtain location information of interest; The information about the places of interest is pushed to the target user.

2. The method according to claim 1, characterized in that, The current location information includes first geographic information and first tag information of the current location. Each candidate location information includes second geographic information and second tag information of the candidate location. Based on each candidate location information and the current location information, multiple location association data are obtained, including: Based on the first geographic information and the second geographic information of each candidate location, the geographic location between each candidate location and the current location is compared to obtain multiple directional association data, and the directional association data corresponds one-to-one with the candidate locations; Based on the first tag information and the second tag information of each candidate location, the category tags between each candidate location and the current location are compared to obtain multiple tag association data, and the tag association data corresponds one-to-one with the candidate location; The location association data is obtained based on the location association data and the tag association data.

3. The method according to claim 1, characterized in that, The step of filtering multiple candidate location information based on the filtering condition data to obtain location information of interest includes: Based on the updated interest feature vector, the location probability is calculated to obtain the interest probability feature of each candidate location; Based on the interest probability features, the interest location information is selected from multiple candidate location information.

4. The method according to any one of claims 1 to 3, characterized in that, The user information includes first geographic information and historical behavior data. The step of obtaining current location information of the target user and candidate location information from multiple candidate locations based on the user information includes: Based on the first geographic information, the current location information is obtained; Based on the current location information and the historical behavior data, information is filtered in a preset location information database to obtain the candidate location information of multiple candidate locations.

5. An electronic device, characterized in that, include: The device includes a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the interest location push method as described in any one of claims 1 to 4.

6. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that is executed by a processor to implement the interest location push method as described in any one of claims 1 to 4.

Images