Method for determining geographical point of interest and method for training geographical point of interest determination model

Abstract

The present disclosure relates to a method for determining a geographical point of interest, a method for training a geographical point of interest determination model, an apparatus, an electronic device, and a computer-readable storage medium. The method for training the geographical point of interest determination model includes: determining an initial geographical point of interest matching the account according to the account attribute information of the account; obtaining the geographical point of interest attribute information corresponding to the initial geographical point of interest; obtaining the access probability of the account for the initial geographical point of interest according to the geographical point of interest attribute information and the account attribute information; and screening out the target geographical point of interest corresponding to the account from the initial geographical points of interest according to the access probability. The method for determining a geographical point of interest provided by the present disclosure can determine a geographical point of interest by combining the account attributes and the attributes of the geographical point of interest, thereby improving the intelligence of the determination of the geographical point of interest.

Description

Technical Field

[0001] The present disclosure relates to the field of Internet technologies, and in particular, to a method for determining a geographical point of interest, a method for training a geographical point of interest determination model, an apparatus, an electronic device, and a computer-readable storage medium. Background Art

[0002] With the development of Internet technologies, in Internet consumption scenarios such as short videos or live broadcasts, technologies for implementing information recommendation based on geographical locations have also developed. Among them, the technology for implementing information recommendation using geographical points of interest is one of the conventional means of such geographical location-based information recommendation technologies. For example, when recommending short videos to an account, store information and coupon information of nearby stores can be recommended to it.

[0003] In related technologies, in the current technology for implementing information recommendation using geographical points of interest, the determination of geographical points of interest mainly adopts the distance-priority determination principle, that is, the geographical points of interest that are closer to the current location of the account are used as the geographical points of interest for which information recommendation is to be performed. Therefore, the current determination of geographical points of interest is relatively simple and not intelligent enough. Summary of the Invention

[0004] The present disclosure provides a method for determining a geographical point of interest, a method for training a geographical point of interest determination model, an apparatus, an electronic device, and a computer-readable storage medium, so as to at least solve the problem that the determination of geographical points of interest in related technologies is relatively simple and not intelligent enough. The technical solution of the present disclosure is as follows:

[0005] According to a first aspect of an embodiment of the present disclosure, a method for determining a geographical point of interest is provided, including:

[0006] Determine an initial geographical point of interest that matches the account according to the account attribute information of the account;

[0007] Obtain geographical point of interest attribute information corresponding to the initial geographical point of interest;

[0008] Input the trained geographical point of interest determination model according to the geographical point of interest attribute information and the account attribute information to obtain the access probability of the account for the initial geographical point of interest;

[0009] Screen out a target geographical point of interest corresponding to the account from the initial geographical points of interest according to the access probability.

[0010] In an exemplary embodiment, obtaining the access probability of the account for the initial geographical point of interest according to the geographical point of interest attribute information and the account attribute information includes: inputting the geographical point of interest attribute information and the account attribute information into a trained geographical point of interest determination model to obtain the access probability; the geographical point of interest determination model is obtained by training a geographical point of interest determination model to be trained according to the sample account attribute information of the sample account, the sample geographical point of interest attribute information of the sample geographical point of interest, and the actual access probability of the sample account for the sample geographical point of interest.

[0011] In an exemplary embodiment, determining an initial geographical point of interest matching the account according to the account attribute information of the account includes: obtaining candidate geographical points of interest according to the account attribute information; obtaining candidate interest point attribute information corresponding to the candidate geographical points of interest; and determining the initial geographical point of interest from the candidate geographical points of interest based on the candidate interest point attribute information.

[0012] In an exemplary embodiment, the account attribute information includes the account location of the account; obtaining candidate geographical points of interest according to the account attribute information includes: using geographical points of interest whose distance from the account location is less than a preset distance threshold as the candidate geographical points of interest.

[0013] In an exemplary embodiment, after screening out the target geographical points of interest corresponding to the account from the initial geographical points of interest, it further includes: obtaining recommendation information associated with the target geographical points of interest according to the access probability of the target geographical points of interest; and pushing the recommendation information to the account.

[0014] In an exemplary embodiment, obtaining recommendation information associated with the target geographical points of interest according to the access probability of the target geographical points of interest includes: obtaining recommendation order information of the target geographical points of interest according to the access probability of the target geographical points of interest; and obtaining recommendation information associated with the target geographical points of interest according to the recommendation order information.

[0015] According to a second aspect of the embodiments of the present disclosure, a method for training a geographical point of interest determination model is provided, including:

[0016] Obtaining sample account attribute information of a sample account, sample geographical point of interest attribute information of a sample geographical point of interest, and the actual access probability of the sample account for the sample geographical point of interest;

[0017] Extracting sample account features corresponding to the sample account attribute information and sample interest point features corresponding to the sample geographical point of interest attribute information;

[0018] Input the sample account features and the sample point-of-interest features into a geographical point-of-interest determination model to be trained, and obtain the predicted access probability of the sample account for the sample geographical point of interest;

[0019] Train the geographical point-of-interest determination model to be trained according to the actual access probability and the predicted access probability to obtain the trained geographical point-of-interest determination model.

[0020] In an exemplary embodiment, after obtaining the trained geographical point-of-interest determination model, the method further includes: obtaining verification data for model tuning of the trained geographical point-of-interest determination model from the sample account attribute information and the sample geographical point-of-interest attribute information; inputting the verification data into the trained geographical point-of-interest determination model, and using the actual access probability corresponding to the verification data to perform model tuning on the trained geographical point-of-interest determination model.

[0021] In an exemplary embodiment, after performing model tuning on the trained geographical point-of-interest determination model, the method further includes: obtaining test data for testing the geographical point-of-interest determination model after model tuning from the sample account attribute information and the sample geographical point-of-interest attribute information; inputting the test data into the geographical point-of-interest determination model after model tuning, and using the actual access probability corresponding to the test data to test the geographical point-of-interest determination model after model tuning.

[0022] According to a third aspect of the embodiments of the present disclosure, there is provided a device for determining a geographical point of interest, including:

[0023] An initial point-of-interest determination unit configured to determine an initial geographical point of interest matching the account according to the account attribute information of the account;

[0024] A point-of-interest attribute acquisition unit configured to acquire geographical point-of-interest attribute information corresponding to the initial geographical point of interest;

[0025] An access probability acquisition unit configured to obtain the access probability of the account for the initial geographical point of interest according to the geographical point-of-interest attribute information and the account attribute information;

[0026] A target point-of-interest determination unit configured to screen out a target geographical point of interest corresponding to the account from the initial geographical points of interest according to the access probability.

[0027] In an exemplary embodiment, the access probability obtaining unit is further configured to input the geographical point of interest attribute information and the account attribute information into a trained geographical point of interest determination model to obtain the access probability; the geographical point of interest determination model is obtained by training a geographical point of interest determination model to be trained according to the sample account attribute information of the sample account, the sample geographical point of interest attribute information of the sample geographical point of interest, and the actual access probability of the sample account for the sample geographical point of interest.

[0028] In an exemplary embodiment, the initial point of interest determination unit is further configured to obtain candidate geographical points of interest according to the account attribute information; obtain candidate interest point attribute information corresponding to the candidate geographical points of interest; and determine the initial geographical point of interest from the candidate geographical points of interest based on the candidate interest point attribute information.

[0029] In an exemplary embodiment, the account attribute information includes the account location of the account; the initial point of interest determination unit is further configured to use geographical points of interest whose distance from the account location is less than a preset distance threshold as the candidate geographical points of interest.

[0030] In an exemplary embodiment, the geographical point of interest determination device further includes: a recommendation information pushing unit configured to obtain recommendation information associated with the target geographical point of interest according to the access probability of the target geographical point of interest; and push the recommendation information to the account.

[0031] In an exemplary embodiment, the recommendation information pushing unit is further configured to obtain recommendation order information of the target geographical point of interest according to the access probability of the target geographical point of interest; and obtain recommendation information associated with the target geographical point of interest according to the recommendation order information.

[0032] According to a fourth aspect of the embodiments of the present disclosure, there is provided a training device for a geographical point of interest determination model, including:

[0033] A sample information obtaining unit configured to obtain sample account attribute information of a sample account, sample geographical point of interest attribute information of a sample geographical point of interest, and the actual access probability of the sample account for the sample geographical point of interest;

[0034] A sample feature extraction unit configured to extract sample account features corresponding to the sample account attribute information and sample interest point features corresponding to the sample geographical point of interest attribute information;

[0035] A predicted probability obtaining unit, configured to input the sample account features and the sample point-of-interest features into a geographical point-of-interest determination model to be trained, and obtain a predicted access probability of the sample account for the sample geographical point-of-interest;

[0036] A model training unit, configured to train the geographical point-of-interest determination model to be trained according to the actual access probability and the predicted access probability, and obtain the trained geographical point-of-interest determination model.

[0037] In an exemplary embodiment, the training apparatus for the geographical point-of-interest determination model further includes: a model tuning module, configured to obtain verification data for tuning the trained geographical point-of-interest determination model from the sample account attribute information and the sample geographical point-of-interest attribute information; input the verification data into the trained geographical point-of-interest determination model, and use the actual access probability corresponding to the verification data to tune the trained geographical point-of-interest determination model.

[0038] In an exemplary embodiment, the training apparatus for the geographical point-of-interest determination model further includes: configured to obtain test data for testing the geographical point-of-interest determination model after model tuning from the sample account attribute information and the sample geographical point-of-interest attribute information; input the test data into the geographical point-of-interest determination model after model tuning, and use the actual access probability corresponding to the test data to test the geographical point-of-interest determination model after model tuning.

[0039] According to a fifth aspect of the embodiments of the present disclosure, there is provided an electronic device, including: a processor; a memory for storing instructions executable by the processor; wherein, the processor is configured to execute the instructions to implement the method for determining a geographical point-of-interest according to any one of the embodiments in the first aspect, or the method for training a geographical point-of-interest determination model according to any one of the embodiments in the second aspect.

[0040] According to a fourth aspect of the embodiments of the present disclosure, there is provided a computer-readable storage medium, characterized in that when instructions in the computer-readable storage medium are executed by a processor of an electronic device, the electronic device is enabled to execute the method for determining a geographical point-of-interest according to any one of the embodiments in the first aspect, or the method for training a geographical point-of-interest determination model according to any one of the embodiments in the second aspect.

[0041] According to a fifth aspect of the embodiments of the present disclosure, there is provided a computer program product including instructions which, when executed by a processor of an electronic device, enable the electronic device to execute the method for determining a geographical point of interest according to any one of the embodiments in the first aspect, or the method for training a geographical point of interest determination model according to any one of the embodiments in the second aspect.

[0042] The technical solutions provided by the embodiments of the present disclosure at least bring the following beneficial effects:

[0043] By determining an initial geographical point of interest matching the account according to the account attribute information of the account; obtaining the geographical point of interest attribute information corresponding to the initial geographical point of interest; obtaining the access probability of the account for the initial geographical point of interest according to the geographical point of interest attribute information and the account attribute information; and screening out the target geographical point of interest corresponding to the account from the initial geographical points of interest. The present disclosure can screen out the target geographical points of interest that can be used for information recommendation through the account attribute of the account and the attribute of the initial geographical point of interest corresponding to the account. Compared with directly using the geographical points of interest close to the account location as the geographical points of interest for information recommendation, the method for determining geographical points of interest provided by the present disclosure can determine geographical points of interest by combining the account attribute and the attribute of the geographical point of interest, thereby improving the intelligence of determining geographical points of interest.

[0044] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and cannot limit the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] The accompanying drawings herein are incorporated into the specification and constitute a part of the specification, showing embodiments consistent with the present disclosure and used together with the specification to explain the principles of the present disclosure, and do not constitute an improper limitation of the present disclosure.

[0046] Figure 1 is a flowchart of a method for determining a geographical point of interest shown according to an exemplary embodiment.

[0047] Figure 2 is a flowchart of determining an initial geographical point of interest matching the account shown according to an exemplary embodiment.

[0048] Figure 3 is a flowchart of a method for training a geographical point of interest determination model shown according to an exemplary embodiment.

[0049] Figure 4 is a flowchart of model tuning for a geographical point of interest determination model shown according to an exemplary embodiment.

[0050] Figure 5It is a flowchart for model testing of a geographical point of interest determination model shown according to an exemplary embodiment.

[0051] Figure 6 It is a flowchart for model training in a geographical point of interest recommendation method shown according to an exemplary embodiment.

[0052] Figure 7 It is a flowchart for online application in a geographical point of interest recommendation method shown according to an exemplary embodiment.

[0053] Figure 8 It is a block diagram of a device for determining a geographical point of interest shown according to an exemplary embodiment.

[0054] Figure 9 It is a block diagram of a training device for a geographical point of interest determination model shown according to an exemplary embodiment.

[0055] Figure 10 It is a block diagram of an electronic device shown according to an exemplary embodiment. Detailed implementation manners

[0056] In order to enable those of ordinary skill in the art to better understand the technical solutions of the present disclosure, the technical solutions in the embodiments of the present disclosure will be clearly and completely described below with reference to the accompanying drawings.

[0057] It should be noted that the terms "first", "second", etc. in the specification and claims of the present disclosure and the above-mentioned drawings are used to distinguish similar objects, and do not necessarily need to describe a specific order or sequence. It should be understood that such used data can be interchanged under appropriate circumstances so that the embodiments of the present disclosure described herein can be implemented in an order other than those illustrated or described herein. The implementation manners described in the following exemplary embodiments do not represent all implementation manners consistent with the present disclosure. On the contrary, they are merely examples of devices and methods consistent with some aspects of the present disclosure as detailed in the appended claims.

[0058] It should also be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for display, data for analysis, etc.) involved in the present disclosure are all information and data authorized by the user or fully authorized by all parties.

[0059] Figure 1 It is a flowchart of a method for determining a geographical point of interest shown according to an exemplary embodiment. As Figure 1 shown, the method for determining a geographical point of interest can be used in a terminal and includes the following steps.

[0060] In step S101, based on the account attribute information of the account, determine the initial geographical point of interest that matches the account.

[0061] Among them, the initial geographical point of interest is a geographical point of interest determined in advance according to the account attribute of the account. The account can be an account for which recommendation information related to the geographical point of interest needs to be recommended. The account attribute information refers to the account attributes of the account for which the recommendation information needs to be recommended. For example, it can include the current location of the account, the type of geographical point of interest that the account tends to browse, which can be a geographical point of interest with a tendency for dining type, shopping type, or scenic spot type, and whether the account has a preference for browsing geographical points of interest and other related account attributes. Specifically, when the terminal needs to recommend recommendation information carrying geographical point of interest related information to the account, it can first collect the account attribute information of the account, and based on the account attribute information of the account, determine a geographical point of interest that matches the account attribute of the account as the initial geographical point of interest.

[0062] For example, when an account tends to browse geographical points of interest of the dining type, then the terminal can determine the geographical points of interest of the dining type as the initial geographical points of interest from various geographical points of interest. If an account tends to browse geographical points of interest of the shopping type, the terminal can screen out the geographical points of interest of the shopping type from various geographical points of interest as the initial geographical points of interest. In addition, it can also be to determine the initial geographical point of interest according to the location of the account, that is, the geographical points of interest near the current location of the account can also be used as the initial geographical points of interest.

[0063] Step S102, obtain the geographical point of interest attribute information corresponding to the initial geographical point of interest.

[0064] The geographical point of interest attribute information refers to the attribute information corresponding to the initial geographical point of interest obtained in step S101. For example, it can include: the location of the point of interest of the initial geographical point of interest, the type of the initial geographical point of interest, the number of visits to the initial geographical point of interest, and the information completeness of the initial geographical point of interest, etc. Specifically, after the terminal determines the initial geographical point of interest, it can obtain the attribute information corresponding to the initial geographical point of interest as the geographical point of interest attribute information corresponding to the initial geographical point of interest.

[0065] Step S103, based on the geographical point of interest attribute information and the account attribute information, obtain the access probability of the account for the initial geographical point of interest;

[0066] Step S104, according to the access probability, screen out the target geographical points of interest corresponding to the account from the initial geographical points of interest.

[0067] The access probability can be used to characterize the likelihood of the account accessing the initial geographical point of interest. The greater the access probability, the greater the likelihood of the account accessing the initial geographical point of interest. The target geographical point of interest refers to the finally determined geographical point of interest, which can be used to provide relevant recommendation information for the account. Specifically, after the terminal obtains the geographical point of interest attribute information corresponding to the initial geographical point of interest, it can obtain the probability of the user accessing the initial geographical point of interest based on the account attribute information of the account and the geographical point of interest attribute information. For example, through a trained geographical point of interest determination model, the geographical point of interest attribute information and the account attribute information are processed to obtain the access probability of the account for the initial geographical point of interest. Another example is that the access situations of accounts with different account attributes for the selected initial geographical points of interest can be pre-collected, and an account that matches the account attribute of the account for which geographical point of interest recommendation is required can be found from the access situations. After obtaining the access probability of the account based on the access situation of the matching account for the initial geographical point of interest, the target geographical point of interest finally used for information recommendation for the account can be screened out from the initial geographical points of interest. For example, the initial geographical point of interest with an access probability greater than a pre-set access probability threshold can be used as the target geographical point of interest.

[0068] For example, the access situations of account A, account B, account C, and account D for a certain initial geographical point of interest A can be pre-collected. Among them, the account attributes of account A and account D match the account attribute information of account E for which geographical point of interest-related recommendation information is to be recommended. Account A accessed the initial geographical point of interest A, while account D did not access the initial geographical point of interest A. Then, the access probability of account E for the initial geographical point of interest A can be obtained as 50%.

[0069] In the above method for determining the geographical point of interest, the initial geographical point of interest matching the account is determined according to the account attribute information of the account; the geographical point of interest attribute information corresponding to the initial geographical point of interest is obtained; the access probability of the account for the initial geographical point of interest is obtained according to the geographical point of interest attribute information and the account attribute information; and the target geographical point of interest corresponding to the account is screened out from the initial geographical points of interest according to the access probability. Through the account attribute of the account and the attribute of the initial geographical point of interest corresponding to the account, the present disclosure can further screen out the target geographical point of interest that can be used for information recommendation. Compared with directly using the geographical point of interest close to the account location as the geographical point of interest for information recommendation, the method for determining the geographical point of interest provided by the present disclosure can determine the geographical point of interest by combining the account attribute and the attribute of the geographical point of interest, thereby improving the intelligence of the determination of the geographical point of interest.

[0070] In an exemplary embodiment, step S103 may further include: inputting the geographical interest point attribute information and the account attribute information into the trained geographical interest point determination model to obtain an access probability; the geographical interest point determination model is obtained by training the geographical interest point determination model to be trained according to the sample account attribute information of the sample account, the sample geographical interest point attribute information of the sample geographical interest point, and the actual access probability of the sample account for the sample geographical interest point.

[0071] The geographical interest point determination model is a pre-trained network model for determining the access probability of the account for the initial geographical interest point. This model is trained by the pre-collected sample account attribute information, sample geographical interest point attribute information, and the actual access probability of the sample account for the sample geographical interest point. Among them, the sample account attribute information refers to the account attribute information of the account used to train the geographical interest point determination model, that is, the account attribute information of the sample account, while the sample geographical interest point attribute information is the attribute information of the geographical interest point used to train the geographical interest point determination model, that is, the attribute information of the sample geographical interest point. The actual access probability is the access probability of the sample account for the sample geographical interest point, which is used to represent whether the sample user has visited the sample geographical interest point. This access probability can be displayed in the form of a label. Specifically, the terminal can pre-collect the sample account attribute information of the sample account, the sample geographical interest point attribute information of the sample geographical interest point, and the actual access probability of the sample account for the sample geographical interest point, so as to use the above information to train the geographical interest point determination model to be trained, and then obtain the trained geographical interest point determination model. After that, when it is necessary to obtain the access probability of the account for the initial geographical interest point, the account attribute information and the geographical interest point attribute information of the account can be input into the pre-trained geographical interest point determination model, and the geographical interest point determination model outputs the access probability of the account for each initial geographical interest point.

[0072] In this embodiment, the access probability of the account for each initial geographical interest point can be obtained through the pre-trained geographical interest point determination model. This geographical interest point determination model is trained according to the sample account attribute information, sample geographical interest point attribute information, and the actual access probability of the sample account for the sample geographical interest point, so as to improve the accuracy of the obtained access probability of the account for the initial geographical interest point.

[0073] In an exemplary embodiment, as Figure 2 shown, step S101 may further include:

[0074] In step S201, candidate geographical interest points are obtained according to the account attribute information.

[0075] In this embodiment, in order to prevent the excessive amount of geographic interest point attribute information of the initial geographic interest point determined by directly using the account attribute information to obtain the input geographic interest point determination model, which may affect the operation speed of the geographic interest point determination model, the terminal in this embodiment can first perform a preliminary screening on the geographic interest points filtered out by the account attribute information, that is, the candidate geographic interest points, and then obtain the initial geographic interest points for inputting the geographic interest point attribute information into the geographic interest point determination model. Specifically, the terminal can first determine, from multiple geographic interest points, the geographic interest points that are adapted to the account attribute information according to the account attribute information of the account as candidate geographic interest points.

[0076] In step S202, obtain the candidate interest point attribute information corresponding to the candidate geographic interest points.

[0077] The candidate interest point attribute information refers to the attribute information of the candidate geographic interest points filtered out by the terminal in step S201. After the terminal determines the candidate geographic interest points, it can obtain the attribute information of the candidate geographic interest points as the candidate interest point attribute information.

[0078] In step S203, determine the initial geographic interest points from the candidate geographic interest points based on the candidate interest point attribute information.

[0079] Among them, determining the initial geographic interest points from the candidate geographic interest points can be performed by screening based on a pre-set geographic interest point screening rule. For example, the rule can be a screening rule based on the access times of each candidate geographic interest point, that is, screening out the candidate geographic interest points with relatively more access times from the obtained candidate geographic interest points as the initial geographic interest points. The rule can also be a screening rule based on the account praise rate of each candidate geographic interest point, that is, screening out the candidate geographic interest points with relatively higher praise rates from the obtained candidate geographic interest points as the initial geographic interest points. Or it can be a screening rule based on the location of each candidate geographic interest point, that is, screening out the candidate geographic interest points with locations relatively close to the current location of the account from the obtained candidate geographic interest points as the initial geographic interest points, and so on.

[0080] In this embodiment, the terminal can screen the candidate geographic interest points obtained according to the account attribute information and then obtain the initial geographic interest points, which can reduce the amount of geographic interest point attribute information of the initial geographic interest points input into the geographic interest point determination model, thereby improving the operation speed of the geographic interest point determination model.

[0081] Furthermore, the account attribute information may include the account location of the account; step S201 may further include: taking the geographic interest points whose distance from the account location is less than a preset distance threshold as candidate geographic interest points.

[0082] In this embodiment, the selection of candidate points of interest (POIs) can be determined according to the account location of the account in the account attribute information. The terminal can filter out the POIs close to the account location from multiple POIs as candidate POIs. Specifically, the terminal can determine the current account location of the account from the account attribute information, and use a pre-set distance threshold to take the POIs whose distance from the current account location is less than the distance threshold as candidate POIs.

[0083] In this embodiment, the terminal can take the POIs whose distance from the account location is less than the pre-set distance threshold as candidate POIs, so as to ensure that the location of the target POI finally selected by the POI determination model can be near the current account location of the account, thereby further improving the relevance between the determined target POI and the account.

[0084] In an exemplary embodiment, after step S103, it may further include: obtaining recommendation information associated with the target POI according to the access probability of the target POI; pushing the recommendation information to the account.

[0085] After the terminal obtains the target POI, it can also obtain the recommendation information associated with the target POI based on the determined target POI. For example, if the determined target POI is a certain catering enterprise, then the terminal can obtain the information associated with the catering enterprise, which can be the relevant preferential information of the catering enterprise as the recommendation information associated with the target POI and push it to the account. If the determined recommendation information associated with the target POI is a certain shopping store, then the terminal can obtain the merchant information of the store as the recommendation information associated with the target POI and push it to the account.

[0086] Specifically, after the terminal obtains the target POI, it can find the recommendation information associated with the target POI based on the access probability of the determined target POI, which can be to select the recommendation information associated with the target POI with the highest access probability from multiple recommendation letters related to the target POI and push it to the account.

[0087] In this embodiment, the terminal can also obtain the recommendation information associated with the target POI based on the access probability corresponding to the determined target POI and push it to the account, which can make the recommendation information recommended by the terminal more in line with the needs of the account and increase the access probability of the account to the recommendation information.

[0088] Further, obtaining recommendation information associated with the target geographical point of interest according to the access probability of the target geographical point of interest may further include: obtaining the recommendation order information of the target geographical point of interest according to the access probability of the target geographical point of interest; and obtaining the recommendation information associated with the target geographical point of interest according to the recommendation order information.

[0089] Among them, the recommendation order information refers to the size order information of the access probability of the account for the target geographical point of interest. After the terminal obtains the access probability of the target geographical point of interest, it can sort the sizes of the access probabilities of the target geographical points of interest, so as to obtain the recommendation order information corresponding to each target geographical point of interest. After that, the terminal can obtain the recommendation information associated with the target geographical point of interest based on the above recommendation order information. For example, after the terminal obtains the recommendation order information corresponding to the target geographical point of interest, it can use the above recommendation order information as an input dimension of a preset recommendation engine, and use the recommendation engine to output the above recommendation information.

[0090] In this embodiment, the recommendation information can be obtained based on the access probability of the target geographical point of interest, which can make the recommendation information for pushing information to the account be the recommendation information that best meets the needs of the account, thereby improving the recommendation quality of the recommendation information and the traffic conversion rate of the recommendation information.

[0091] Figure 3 It is a flowchart of a method for training a geographical point of interest determination model shown according to an exemplary embodiment. As Figure 3 shown, the method for training a geographical point of interest determination model can be used in a terminal and includes the following steps.

[0092] In step S301, obtain the sample account attribute information of the sample account, the sample geographical point of interest attribute information of the sample geographical point of interest, and the actual access probability of the sample account for the sample geographical point of interest.

[0093] Among them, the sample account attribute information refers to the account attribute information of the account collected in advance by the terminal for training the geographical point of interest determination model, that is, the account attribute information of the sample account. The sample geographical point of interest attribute information is the attribute information of the geographical point of interest collected in advance by the terminal for training the geographical point of interest determination model, that is, the attribute information of the sample geographical point of interest. The actual access probability refers to the access probability of the sample account for the sample geographical point of interest, and this access probability can be displayed in the form of a label. For example, if the sample account has actually visited the sample geographical point of interest, then the actual access probability of the sample account for the sample geographical point of interest can be set to 1, and if the sample account has not actually visited the sample geographical point of interest, then the actual access probability of the sample account for the sample geographical point of interest can be set to 0.

[0094] Specifically, the terminal can pre-collect historical access information of a sample account accessing a sample geographical point of interest, and can determine the actual access probability of the sample account for the sample geographical point of interest based on the historical access information, as well as the account attribute information of the sample account and the attribute information of the sample geographical point of interest, respectively as the sample account attribute information and the sample geographical point of interest attribute information.

[0095] In step S302, extract the sample account features corresponding to the sample account attribute information and the sample point of interest features corresponding to the sample geographical point of interest attribute information.

[0096] The sample account features refer to the account features obtained by extracting features from the sample account attribute information, while the sample point of interest features refer to the point of interest features obtained by extracting features from the sample geographical point of interest attribute information. The terminal can perform data cleaning and transformation on the sample account attribute information and the sample geographical point of interest attribute information obtained in step S301, so as to obtain the sample account features and the sample point of interest features.

[0097] In step S303, input the sample account features and the sample point of interest features into the geographical point of interest determination model to be trained, and obtain the predicted access probability of the sample account for the sample geographical point of interest;

[0098] In step S304, train the geographical point of interest determination model to be trained according to the actual access probability and the predicted access probability, and obtain the trained geographical point of interest determination model.

[0099] The geographical point of interest determination model to be trained can be a geographical point of interest determination model that needs to be trained. The predicted access probability refers to the probability that the geographical point of interest determination model to be trained predicts that the sample account accesses the sample geographical point of interest based on the input sample account features and sample point of interest features. The terminal can input the sample account features and the sample point of interest features into the geographical point of interest determination model that needs to be trained. The geographical point of interest determination model predicts the probability that the sample account accesses the sample geographical point of interest, that is, outputs the predicted access probability of the sample account for the sample geographical point of interest. Then, the terminal can use the actual access probability and the predicted access probability of the sample account for the sample geographical point of interest as the difference loss of the model, and then train the geographical point of interest determination model, so as to obtain the trained geographical point of interest determination model.

[0100] In the training method of the above-mentioned geographical point of interest determination model, by obtaining the sample account attribute information of the sample account, the sample geographical point of interest attribute information of the sample geographical point of interest, and the actual access probability of the sample account for the sample geographical point of interest; extracting the sample account features corresponding to the sample account attribute information and the sample interest point features corresponding to the sample geographical point of interest attribute information; inputting the sample account features and the sample interest point features into the geographical point of interest determination model to be trained to obtain the predicted access probability of the sample account for the sample geographical point of interest; and training the geographical point of interest determination model to be trained according to the actual access probability and the predicted access probability to obtain the trained geographical point of interest determination model. The present disclosure can train the geographical point of interest determination model through the pre-collected sample account attribute information, sample geographical point of interest attribute information, and the actual access probability of the sample account for the sample geographical point of interest, so as to obtain the trained geographical point of interest determination model, and realize the determination of the geographical point of interest by using the geographical point of interest determination model, thereby improving the intelligence of the determined geographical point of interest.

[0101] In addition, as Figure 4 shown, after step S304, it may further include:

[0102] In step S401, test data for testing the optimized geographical point of interest determination model is obtained from the sample account attribute information and the sample geographical point of interest attribute information.

[0103] In this embodiment, after the geographical point of interest determination model is trained, the trained geographical point of interest determination model can be further optimized. Specifically, in this embodiment, after obtaining the sample account attribute information and the sample geographical point of interest attribute information, the sample account attribute information and the sample geographical point of interest attribute information can be classified according to a certain ratio as training data for training the geographical point of interest determination model and verification data for optimizing the trained geographical point of interest determination model. After the geographical point of interest determination model is trained, the corresponding verification data can be obtained from the pre-classified sample account attribute information and sample geographical point of interest attribute information.

[0104] For example, after the terminal completes the collection of the sample account attribute information and the sample geographical point of interest attribute information, the collected sample account attribute information and the sample geographical point of interest attribute information can be divided into training data and verification data according to a preset ratio. After the terminal uses the training data to complete the training of the geographical point of interest determination model, the pre-divided verification data can be obtained from the sample account attribute information and the sample geographical point of interest attribute information.

[0105] In step S402, the verification data is input into the trained geographical point of interest determination model, and the trained geographical point of interest determination model is optimized using the actual access probability corresponding to the verification data.

[0106] After that, the terminal can input the collected verification data into the trained geographical point of interest determination model, so that the predicted access probability corresponding to the verification data can be obtained through the geographical point of interest determination model, and the trained geographical point of interest determination model can be further optimized using the actual access probability corresponding to the verification data and the predicted access probability corresponding to the verification data.

[0107] In this embodiment, after the terminal completes the training of the geographical point of interest determination model, the verification data can be further used to optimize the geographical point of interest determination model, which can further improve the accuracy of the trained geographical point of interest determination model, thereby improving the accuracy of the determined geographical points of interest.

[0108] Further, as Figure 5 shown, after step S402, it may further include:

[0109] In step S501, test data for testing the optimized geographical point of interest determination model is obtained from the sample account attribute information and the sample geographical point of interest attribute information.

[0110] In this embodiment, after the geographical point of interest determination model is optimized, it can be further tested. Specifically, after obtaining the sample account attribute information and the sample geographical point of interest attribute information in this embodiment, in addition to classifying the sample account attribute information and the sample geographical point of interest attribute information according to a certain ratio to obtain training data and verification data, a part of them can be further classified as test data for testing the optimized geographical point of interest determination model. After the geographical point of interest determination model is optimized, the corresponding test data can be obtained from the pre-classified sample account attribute information and the sample geographical point of interest attribute information.

[0111] For example, after the terminal completes the collection of the sample account attribute information and the sample geographical point of interest attribute information, the collected sample account attribute information and the sample geographical point of interest attribute information can be divided into training data, verification data, and test data according to a preset ratio, which can be in the ratio of 7:2:1. After the terminal uses the verification data to optimize the geographical point of interest determination model, the pre-divided test data can be further obtained from the sample account attribute information and the sample geographical point of interest attribute information.

[0112] In step S502, input the test data into the geographical point of interest determination model after model tuning, and use the actual access probability corresponding to the test data to test the geographical point of interest determination model after model tuning.

[0113] After completing the tuning of the geographical point of interest determination model, the terminal can also input the test data into the geographical point of interest determination model after tuning, so that the predicted access probability corresponding to the test data can be obtained through the geographical point of interest determination model, and the overall performance of the trained geographical point of interest determination model can be tested by using the actual access probability corresponding to the test data and the predicted access probability corresponding to the test data.

[0114] In this embodiment, after the terminal completes the model tuning of the geographical point of interest determination model by using the verification data, the test data can also be used to perform further performance testing on the geographical point of interest determination model, which can further ensure the model performance of the geographical point of interest determination model.

[0115] In an exemplary embodiment, a geographical point of interest recommendation method based on multi-dimensional attributes is also provided. Through the request information when the user uses the app and the multi-dimensional information of the POI (i.e., geographical point of interest), many POIs suitable for the user are trained through the data of these two links, and these POIs are sorted, and based on this sorting, recommendations are made for the user to improve the recommendation quality and efficiency.

[0116] Among them, the user-side data may include:

[0117] (1) User's location: The location when the user uses the app;

[0118] (2) User's local life attribute: Whether the user has the attribute of local life (concerned about scenarios such as dining and shopping), that is, whether the user has the preference for browsing POIs (summarize the characteristics through historical clicks and browsing behaviors of various POIs);

[0119] (3) User's local life preference: The preference for the type of local life that the user tends to browse (such as dining, shopping, scenic spots, etc.) (summarize the characteristics through historical clicks and browsing behaviors of various POIs).

[0120] And the data of the POI-side data may include:

[0121] (1) POI location: The administrative division, longitude and latitude location where the POI is located;

[0122] (2) POI type: The type to which the POI belongs (such as dining, hotel, shopping, scenic spot, etc.);

[0123] (3) POI popularity: The number of times the POI is exposed and clicked;

[0124] (4) POI Attribute Richness: Whether attributes such as price, label, business hours, etc. are available.

[0125] Specifically, this embodiment can be composed of the following two processes:

[0126] 1. Model Training: Train the relationship between user information and the POIs they click on through historical data. The process of model training can be as Figure 6 shown, and specifically can include the following steps:

[0127] (1) Historical Data Collection: Through the information reported by users historically. That is, when users browse videos with POIs, the number of clicks / ratio of users under each POI category, the stay duration on the POI details page, etc. are used to depict the local life attributes and local life preferences of users; and based on the location of users each time they browse videos with POIs historically, the category, popularity, information richness, etc. of the POIs hung on the videos, as well as whether the users finally clicked on the POIs hung on the videos.

[0128] (2) Feature Engineering: Clean, transform, etc. the above various data, and split the training data, validation data, and test data.

[0129] (3) Train a Supervised Model: Build a supervised machine learning model through the features and dependent variables (whether the user clicks on the POI) obtained in feature engineering, and optimize the model on the validation set. Finally, obtain the relationship model between user information and POI information when the user browses videos with POIs and whether they click on the POI.

[0130] 2. Online Prediction: Use the obtained model to achieve online prediction. The process of online prediction can be as Figure 7 shown, and specifically can include the following steps:

[0131] (1) POI Recall: According to the location reported by the user when using the APP, find all POIs within a certain range (such as within 1 kilometer) as the recommended candidate set.

[0132] (2) Coarse Sorting: Coarsely sort according to business rules such as POI category, popularity, high value (such as catering, shopping, scenic area categories), and POI popularity.

[0133] (3) Fine Sorting: Select the top 10% of the POIs from the results of the coarse sorting, input the user information and POI information into the trained model to obtain the probability of the user clicking on the POI, and sort the POIs obtained in the coarse sorting from high to low according to the probability. This sorting (or probability value) can be used as a dimension in the video recommendation engine to recommend videos with the POI having the highest prediction probability to users with local life attributes.

[0134] In this embodiment, by combining user data and objective POI data, matching and sorting the two, some more effective POIs (i.e., POIs that users are more likely to like) are obtained. By transmitting the POI information that users may prefer to the video recommendation engine, the conversion and monetization of user online traffic to offline traffic are improved.

[0135] It should be understood that although Figures 1-7 the steps in the flowchart of Figures 1-7 are shown in sequence according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise clearly stated in this article, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover,

[0136] It can be understood that the same / similar parts among the various embodiments of the above methods in this specification can be referred to each other. Each embodiment focuses on the differences from other embodiments. For the relevant parts, refer to the descriptions of other method embodiments.

[0137] Figure 8 is a block diagram of a device for determining a geographical point of interest shown according to an exemplary embodiment. Referring to Figure 8 , the device includes an initial point of interest determination unit 801, a point of interest attribute acquisition unit 802, an access probability acquisition unit 803, and a target point of interest determination unit 804.

[0138] The initial point of interest determination unit 801 is configured to determine an initial geographical point of interest matching the account according to the account attribute information of the account;

[0139] The point of interest attribute acquisition unit 802 is configured to acquire geographical point of interest attribute information corresponding to the initial geographical point of interest;

[0140] The access probability acquisition unit 803 is configured to obtain the access probability of the account for the initial geographical point of interest according to the geographical point of interest attribute information and the account attribute information;

[0141] The target point of interest determination unit 804 is configured to screen out the target geographical point of interest corresponding to the account from the initial geographical points of interest according to the access probability.

[0142] In an exemplary embodiment, the access probability acquisition unit 803 is further configured to input the geographical point of interest attribute information and the account attribute information into the trained geographical point of interest determination model to obtain the access probability; the geographical point of interest determination model is obtained by training the geographical point of interest determination model to be trained according to the sample account attribute information of the sample account, the sample geographical point of interest attribute information of the sample geographical point of interest, and the actual access probability of the sample account for the sample geographical point of interest.

[0143] In an exemplary embodiment, the initial point of interest determination unit 801 is further configured to obtain candidate geographical points of interest according to the account attribute information; obtain candidate interest point attribute information corresponding to the candidate geographical points of interest; and determine an initial geographical point of interest from the candidate geographical points of interest based on the candidate interest point attribute information.

[0144] In an exemplary embodiment, the account attribute information includes the account location of the account; the initial point of interest determination unit 801 is further configured to use the geographical points of interest whose distance from the account location is less than a preset distance threshold as candidate geographical points of interest.

[0145] In an exemplary embodiment, the device for determining a geographical point of interest further includes: a recommendation information pushing unit, configured to obtain recommendation information associated with the target geographical point of interest according to the access probability of the target geographical point of interest; and push the recommendation information to the account.

[0146] In an exemplary embodiment, the recommendation information pushing unit is further configured to obtain the recommendation order information of the target geographical point of interest according to the access probability of the target geographical point of interest; and obtain the recommendation information associated with the target geographical point of interest according to the recommendation order information.

[0147] Figure 9 It is a block diagram of a training device for a geographical point of interest determination model shown according to an exemplary embodiment. Refer to Figure 9 , the device includes a sample information acquisition unit 901, a sample feature extraction unit 902, a predicted probability acquisition unit 903, and a model training unit 904.

[0148] The sample information acquisition unit 901 is configured to obtain the sample account attribute information of the sample account, the sample geographical point of interest attribute information of the sample geographical point of interest, and the actual access probability of the sample account for the sample geographical point of interest;

[0149] The sample feature extraction unit 902 is configured to extract the sample account features corresponding to the sample account attribute information and the sample interest point features corresponding to the sample geographical point of interest attribute information;

[0150] A predicted probability acquisition unit 903 is configured to input sample account features and sample point-of-interest features into a geographical point-of-interest determination model to be trained, and obtain a predicted access probability of the sample account for the sample geographical point-of-interest.

[0151] A model training unit 904 is configured to train the geographical point-of-interest determination model to be trained according to the actual access probability and the predicted access probability, and obtain a trained geographical point-of-interest determination model.

[0152] In an exemplary embodiment, the training device of the geographical point-of-interest determination model further includes: a model tuning module configured to obtain verification data for tuning the trained geographical point-of-interest determination model from sample account attribute information and sample geographical point-of-interest attribute information; input the verification data into the trained geographical point-of-interest determination model, and use the actual access probability corresponding to the verification data to tune the trained geographical point-of-interest determination model.

[0153] In an exemplary embodiment, the training device of the geographical point-of-interest determination model further includes: a model testing module configured to obtain test data for testing the geographical point-of-interest determination model after model tuning from sample account attribute information and sample geographical point-of-interest attribute information; input the test data into the geographical point-of-interest determination model after model tuning, and use the actual access probability corresponding to the test data to test the geographical point-of-interest determination model after model tuning.

[0154] Regarding the device in the above embodiments, the specific manners in which each module performs operations have been described in detail in the embodiments related to the method, and will not be elaborated herein.

[0155] Figure 10 It is a block diagram of an electronic device 1000 for determining a geographical point-of-interest according to an exemplary embodiment. For example, the electronic device 1000 may be a mobile phone, a computer, a digital broadcast terminal, a messaging device, a game console, a tablet device, a medical device, a fitness device, a personal digital assistant, etc.

[0156] Refer to Figure 8 , the electronic device 1000 may include one or more of the following components: a processing component 1002, a memory 1004, a power component 1006, a multimedia component 1008, an audio component 1010, an input / output (I / O) interface 1012, a sensor component 1014, and a communication component 1016.

[0157] The processing component 1002 generally controls the overall operation of the electronic device 1000, such as operations associated with display, telephone calls, data communication, camera operations, and recording operations. The processing component 1002 may include one or more processors 1020 to execute instructions to complete all or part of the steps of the above methods. In addition, the processing component 1002 may include one or more modules to facilitate the interaction between the processing component 1002 and other components. For example, the processing component 1002 may include a multimedia module to facilitate the interaction between the multimedia component 1008 and the processing component 1002.

[0158] The memory 1004 is configured to store various types of data to support the operation of the electronic device 1000. Examples of such data include instructions for any application or method operating on the electronic device 1000, contact data, phone book data, messages, pictures, videos, etc. The memory 1004 may be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk, optical disk, or graphene memory.

[0159] The power component 1006 provides power to various components of the electronic device 1000. The power component 1006 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power for the electronic device 1000.

[0160] The multimedia component 1008 includes a screen that provides an output interface between the electronic device 1000 and the user. In some embodiments, the screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen may be implemented as a touch screen to receive input signals from the user. The touch panel includes one or more touch sensors to sense touches, swipes, and gestures on the touch panel. The touch sensors may sense not only the boundaries of touch or swipe actions but also detect the duration and pressure associated with the touch or swipe operations. In some embodiments, the multimedia component 1008 includes a front camera and / or a rear camera. When the electronic device 1000 is in an operating mode, such as a shooting mode or a video mode, the front camera and / or the rear camera may receive external multimedia data. Each of the front camera and the rear camera may be a fixed optical lens system or have a focal length and optical zoom capabilities.

[0161] The audio component 1010 is configured to output and / or input audio signals. For example, the audio component 1010 includes a microphone (MIC), which is configured to receive external audio signals when the electronic device 1000 is in an operating mode, such as a call mode, a recording mode, and a voice recognition mode. The received audio signals can be further stored in the memory 1004 or transmitted via the communication component 1016. In some embodiments, the audio component 1010 further includes a speaker for outputting audio signals.

[0162] The I / O interface 1012 provides an interface between the processing component 1002 and a peripheral interface module, and the peripheral interface module can be a keyboard, a click wheel, buttons, etc. These buttons can include, but are not limited to: a home button, a volume button, a power button, and a lock button.

[0163] The sensor component 1014 includes one or more sensors for providing status assessments of various aspects of the electronic device 1000. For example, the sensor component 1014 can detect the on / off state of the electronic device 1000, the relative positioning of components, such as the display and keypad of the electronic device 1000. The sensor component 1014 can also detect a change in the position of the electronic device 1000 or an electronic device component, the presence or absence of user contact with the electronic device 1000, the orientation or acceleration / deceleration of the device 1000, and a change in the temperature of the electronic device 1000. The sensor component 1014 can include a proximity sensor configured to detect the presence of nearby objects without any physical contact. The sensor component 1014 can also include a light sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, the sensor component 1014 can further include an acceleration sensor, a gyro sensor, a magnetic sensor, a pressure sensor, or a temperature sensor.

[0164] The communication component 1016 is configured to facilitate communication between the electronic device 1000 and other devices in a wired or wireless manner. The electronic device 1000 can access a wireless network based on a communication standard, such as WiFi, a carrier network (such as 2G, 3G, 4G, or 5G), or a combination thereof. In an exemplary embodiment, the communication component 1016 receives a broadcast signal or broadcast-related information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communication component 1016 further includes a near field communication (NFC) module to facilitate short-range communication. For example, the NFC module can be implemented based on radio frequency identification (RFID) technology, infrared data association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology, and other technologies.

[0165] In an exemplary embodiment, the electronic device 1000 may be implemented by one or more application specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components for performing the above method.

[0166] In an exemplary embodiment, there is also provided a computer-readable storage medium including instructions, such as the memory 1004 including instructions, and the above instructions can be executed by the processor 1020 of the electronic device 1000 to complete the above method. For example, the computer-readable storage medium may be a ROM, a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, and an optical data storage device, etc.

[0167] In an exemplary embodiment, there is also provided a computer program product, and the computer program product includes instructions, and the above instructions can be executed by the processor 1020 of the electronic device 1000 to complete the above method.

[0168] It should be noted that the above-mentioned device, electronic device, computer-readable storage medium, computer program product, etc. may also include other implementation manners according to the description of the method embodiments, and the specific implementation manners can refer to the description of the relevant method embodiments, which will not be elaborated here one by one.

[0169] Those skilled in the art will readily conceive of other embodiments of the present disclosure after considering the specification and practicing the invention disclosed herein. The present disclosure is intended to cover any variations, uses, or adaptations of the present disclosure that follow the general principles of the present disclosure and include known common knowledge or conventional technical means in the technical field not disclosed by the present disclosure. The specification and embodiments are only to be considered as exemplary, and the true scope and spirit of the present disclosure are pointed out by the claims.

[0170] It should be understood that the present disclosure is not limited to the exact structures described above and shown in the drawings, and various modifications and changes can be made without departing from its scope. The scope of the present disclosure is only limited by the appended claims.

Claims

1. A method for determining a geographical point of interest, characterized in that, Including: Determine an initial geographical point of interest matching the account according to the account attribute information of the account; Obtain the geographical point of interest attribute information corresponding to the initial geographical point of interest; Obtain the access probability of the account for the initial geographical point of interest according to the geographical point of interest attribute information and the account attribute information; including: inputting the geographical point of interest attribute information and the account attribute information into a trained geographical point of interest determination model to obtain the access probability; The geographical point of interest determination model is obtained by training a geographical point of interest determination model to be trained according to the sample account attribute information of the sample account, the sample geographical point of interest attribute information of the sample geographical point of interest, and the actual access probability of the sample account for the sample geographical point of interest; Screen out a target geographical point of interest corresponding to the account from the initial geographical points of interest according to the access probability.

2. The method for determining a geographical point of interest according to claim 1, wherein The determining an initial geographical point of interest matching the account according to the account attribute information of the account includes: Obtain candidate geographical points of interest according to the account attribute information; Obtain the candidate interest point attribute information corresponding to the candidate geographical points of interest; Determine the initial geographical point of interest from the candidate geographical points of interest based on the candidate interest point attribute information.

3. The method for determining a geographical point of interest according to claim 2, wherein The account attribute information includes the account location of the account; The obtaining candidate geographical points of interest according to the account attribute information includes: Taking geographical points of interest with a distance less than a preset distance threshold from the account location as the candidate geographical points of interest.

4. The method for determining a geographical point of interest according to claim 1, wherein After screening out the target geographical point of interest corresponding to the account from the initial geographical points of interest, it further includes: Obtain recommendation information associated with the target geographical point of interest according to the access probability of the target geographical point of interest; Push the recommendation information to the account.

5. The method for determining a geographical point of interest according to claim 4, characterized in that, The obtaining recommendation information associated with the target geographical point of interest according to the access probability of the target geographical point of interest includes: Obtain the recommendation order information of the target geographical point of interest according to the access probability of the target geographical point of interest; Obtain recommendation information associated with the target geographical point of interest according to the recommendation order information.

6. A training method for a geographical point of interest determination model, characterized in that, Including: Obtain the sample account attribute information of the sample account, the sample geographical point of interest attribute information of the sample geographical point of interest, and the actual access probability of the sample account for the sample geographical point of interest; Extract the sample account features corresponding to the sample account attribute information and the sample interest point features corresponding to the sample geographical point of interest attribute information; Input the sample account features and the sample interest point features into a geographical point of interest determination model to be trained to obtain the predicted access probability of the sample account for the sample geographical point of interest; Train the geographical point of interest determination model to be trained according to the actual access probability and the predicted access probability to obtain the trained geographical point of interest determination model.

7. The training method of the geographical point of interest determination model according to claim 6, wherein After obtaining the trained geographical point of interest determination model, it further includes: Obtain verification data for model tuning of the trained point of interest determination model from the sample account attribute information and the sample point of interest attribute information; Input the verification data into the trained point of interest determination model, and use the actual access probability corresponding to the verification data to tune the trained point of interest determination model.

8. The training method of the geographical point of interest determination model according to claim 7, characterized in that After tuning the trained point of interest determination model, it further includes: Obtain test data for testing the point of interest determination model after model tuning from the sample account attribute information and the sample point of interest attribute information; Input the test data into the point of interest determination model after model tuning, and use the actual access probability corresponding to the test data to test the point of interest determination model after model tuning.

9. An apparatus for determining a geographical point of interest, characterized in that It includes: An initial point of interest determination unit, configured to determine an initial geographical point of interest matching the account according to the account attribute information of the account; A point of interest attribute acquisition unit, configured to obtain the geographical point of interest attribute information corresponding to the initial geographical point of interest; An access probability acquisition unit, configured to obtain the access probability of the account for the initial geographical point of interest according to the geographical point of interest attribute information and the account attribute information; further configured to input the geographical point of interest attribute information and the account attribute information into the trained point of interest determination model to obtain the access probability; the point of interest determination model is obtained by training the to-be-trained point of interest determination model according to the sample account attribute information of the sample account, the sample geographical point of interest attribute information of the sample geographical point of interest, and the actual access probability of the sample account for the sample geographical point of interest; A target point of interest determination unit, configured to screen out the target geographical point of interest corresponding to the account from the initial geographical points of interest according to the access probability.

10. The determining device for geographical points of interest according to claim 9, characterized in that, The initial point of interest determination unit is further configured to obtain candidate geographical points of interest according to the account attribute information; obtain the candidate interest point attribute information corresponding to the candidate geographical points of interest; and determine the initial geographical points of interest from the candidate geographical points of interest based on the candidate interest point attribute information.

11. The determining device for a geographical point of interest according to claim 10, characterized in that, The account attribute information includes the account location of the account; the initial point of interest determination unit is further configured to use the geographical points of interest whose distance from the account location is less than a preset distance threshold as the candidate geographical points of interest.

12. The determining device for geographical points of interest according to claim 9, wherein It further includes: A recommendation information push unit, configured to obtain recommendation information associated with the target geographical point of interest according to the access probability of the target geographical point of interest; Push the recommendation information to the account.

13. The determining device for geographical points of interest according to claim 12, characterized in that, The recommendation information push unit is further configured to obtain the recommendation order information of the target geographical point of interest according to the access probability of the target geographical point of interest; and obtain the recommendation information associated with the target geographical point of interest according to the recommendation order information.

14. A training device for a geographical point of interest determination model, characterized in that, It includes: A sample information acquisition unit, configured to obtain sample account attribute information of a sample account, sample geographic interest point attribute information of a sample geographic interest point, and an actual access probability of the sample account for the sample geographic interest point; A sample feature extraction unit, configured to extract a sample account feature corresponding to the sample account attribute information and a sample interest point feature corresponding to the sample geographic interest point attribute information; A predicted probability acquisition unit, configured to input the sample account feature and the sample interest point feature into a geographic interest point determination model to be trained, and obtain a predicted access probability of the sample account for the sample geographic interest point; A model training unit, configured to train the geographic interest point determination model to be trained according to the actual access probability and the predicted access probability, and obtain the trained geographic interest point determination model; 15. The training device for the geographical point of interest determination model according to claim 14, wherein Further comprising: A model tuning module, configured to obtain verification data for tuning the trained geographic interest point determination model from the sample account attribute information and the sample geographic interest point attribute information; input the verification data into the trained geographic interest point determination model, and use the actual access probability corresponding to the verification data to tune the trained geographic interest point determination model; 16. The training device for the geographical point of interest determination model according to claim 15, characterized in that, Further comprising: A model testing module, configured to obtain test data for testing the geographic interest point determination model after model tuning from the sample account attribute information and the sample geographic interest point attribute information; input the test data into the geographic interest point determination model after model tuning, and use the actual access probability corresponding to the test data to test the geographic interest point determination model after model tuning; 17. An electronic device, characterized in that, Comprising: A processor; A memory for storing instructions executable by the processor; Wherein, the processor is configured to execute the instructions to implement the method for determining a geographic interest point according to any one of claims 1 to 5, or the method for training a geographic interest point determination model according to any one of claims 6 to 8; 18. A computer-readable storage medium, characterized in that, When the instructions in the computer-readable storage medium are executed by a processor of an electronic device, the electronic device is enabled to execute the method for determining a geographic interest point according to any one of claims 1 to 5, or the method for training a geographic interest point determination model according to any one of claims 6 to 8; 19. A computer program product, the computer program product includes instructions, characterized in that, When the instructions are executed by a processor of an electronic device, the electronic device is enabled to execute the method for determining a geographic interest point according to any one of claims 1 to 5, or the method for training a geographic interest point determination model according to any one of claims 6 to 8.

Images