Method, apparatus, device, and storage medium for map search sorting

Abstract

The present application discloses a method, apparatus, device, and storage medium for map search sorting, belonging to the field of sorting technology. The method includes: obtaining a search request, determining the visual position order of the map based on the search request; obtaining a search quality index that conforms to the map form based on the visual position order, where the search quality index is used to indicate the accuracy of the map search sorting; obtaining a map search sorting model based on the search quality index, and performing map search sorting through the map search sorting model. By obtaining a search quality index that conforms to the map form, the present application makes the training process more in line with the actual situation when training the map search sorting model based on the search quality index, improves the performance of the trained map search sorting model. Therefore, when performing map search sorting based on the map search sorting model, the obtained sorting result has a high accuracy rate.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of sorting technology, and particularly to a method, device, equipment and storage medium for map search sorting. Background Art

[0002] With the development of the Internet, when a user searches for keywords on a map interface through the Internet to obtain the required information, a large amount of information is often obtained. In the face of a large amount of information, a method for map search sorting is needed to help the user quickly find the information with the highest relevance to the keywords from the large amount of information.

[0003] In the related art, the commonly used map search sorting method is to construct a map search sorting model by using the document list method and perform search sorting by using the map search sorting model. In the process of constructing the map search sorting model, NDCG (Normalized Discounted Cumulative Gain) is introduced into the loss function, so that when calculating the output loss by using the loss function, the relative position of the sorted samples will also affect the output loss, that is, when calculating the output loss, the position of the samples arranged in descending order of relevance will be considered, and finally the map search sorting model is trained based on the output loss.

[0004] In the above map search sorting method, the relative position considered during the training of the map search sorting model is the position when all samples are arranged in descending order of relevance. However, as a special form, since the position of each sample on the map is fixed and the samples cannot be arranged in descending order of relevance, the calculation method of the output loss in the related art is obviously not applicable, and the performance of the map search sorting model finally trained based on this loss function is poor. Summary of the Invention

[0005] The embodiments of the present application provide a method, device, equipment and storage medium for map search sorting, which can be used to solve the problems in the related art. The technical solutions are as follows:

[0006] On the one hand, the embodiments of the present application provide a method for map search sorting, and the method includes:

[0007] Obtain a search request, and determine the visual position order of the map based on the search request;

[0008] Obtain a search quality index that conforms to the map form based on the visual position order, and the search quality index is used to indicate the accuracy of search sorting;

[0009] Obtain a map search sorting model based on the search quality index, and perform map search sorting through the map search sorting model.

[0010] In a possible implementation, determining the visual position order of the map based on the search request includes:

[0011] Determining the starting point of the visual position order;

[0012] Based on the starting point and the search request, determining a standard visual radius;

[0013] Determining a sorting strategy for the same visual distance interval, and determining the visual position order of the map based on the standard visual radius and the sorting strategy for the same visual distance interval.

[0014] In a possible implementation, obtaining a search quality metric that conforms to the map form based on the visual position order includes:

[0015] Obtaining a first data set and a second data set, where the first data set is the click-through rate of a first sample under different views, and the second data set is the evaluation data of multiple second samples under the same view. The positions of the first sample and the multiple second samples are both within a reference range, and the reference range is determined based on the search request;

[0016] Based on the first data set, the second data set, and the visual position order, determining a target position loss function that conforms to the map form, where the target position loss function is used to calculate the position loss under the map form;

[0017] Based on the target position loss function, determining the search quality metric that conforms to the map form.

[0018] In a possible implementation, the search quality metric is:

[0019]

[0020] Wherein, the IDCG is the cumulative gain of losses in an ideal situation. The ideal situation is that the training samples are arranged in order according to the relevance to the search request. k represents the sorting position of the training sample, G(k) represents the gain of the training sample, η(k) represents the position loss of the training sample, the position loss is determined based on the visual position order, and the NDCG DisView represents the search quality metric.

[0021] In a possible implementation, obtaining a map search ranking model based on the search quality metric includes:

[0022] Based on the search quality metric, obtaining an expression for the first gradient, where the expression for the first gradient is used to determine the training direction and intensity;

[0023] The map search ranking model is trained based on the expression of the first gradient.

[0024] In a possible implementation, the expression of the first gradient is:

[0025]

[0026] wherein, the i and the k represent the positions of the training samples, the S i represents the relevance score of the training sample i, the S k represents the relevance score of the training sample k, the Δ NDCG represents the change amount of the search quality index after swapping the positions of the training sample i and the training sample k, the σ is a constant, the e is a constant, the C is a loss function, the λ ij is the first gradient, and the is used to indicate differentiation.

[0027] On the other hand, a device for map search ranking is provided, and the device includes:

[0028] A determination module, configured to obtain a search request and determine the visual position order of the map based on the search request;

[0029] An acquisition module, configured to obtain a search quality index conforming to the map form based on the visual position order, and the search quality index is used to indicate the accuracy of the search ranking;

[0030] A ranking module, configured to obtain a map search ranking model based on the search quality index and perform map search ranking through the map search ranking model.

[0031] In a possible implementation, the determination module is configured to determine the starting point of the visual position order; determine a standard visual radius based on the starting point and the search request; determine a ranking strategy in the same visual distance interval, and determine the visual position order of the map based on the standard visual radius and the ranking strategy in the same visual distance interval.

[0032] In a possible implementation, the obtaining module is configured to obtain a first data set and a second data set. The first data set is the click-through rate of a first sample under different views, and the second data set is the evaluation data of multiple second samples under the same view. The positions of the first sample and the multiple second samples are both within a reference range, and the reference range is determined based on the search request. A target position loss function that conforms to the map form is determined based on the first data set, the second data set, and the visual position order. The target position loss function is used to calculate the position loss in the map form. The search quality metric that conforms to the map form is determined based on the target position loss function.

[0033] In a possible implementation, the search quality metric is:

[0034]

[0035] Wherein, the IDCG is the cumulative gain of losses in the ideal case. The ideal case is that the training samples are arranged in order according to their relevance to the search request. k represents the sorting position of the training sample, G(k) represents the gain of the training sample, η(k) represents the position loss of the training sample, and the position loss is determined based on the visual position order. The NDCG DisView represents the search quality metric.

[0036] In a possible implementation, the sorting module is configured to obtain an expression of a first gradient based on the search quality metric. The expression of the first gradient is used to determine the training direction and intensity. The map search sorting model is trained based on the expression of the first gradient.

[0037] In a possible implementation, the expression of the first gradient is:

[0038]

[0039] Wherein, i and k represent the positions of the training samples, S i represents the relevance score of training sample i, S k represents the relevance score of training sample k, Δ NDCG represents the change in the search quality metric after swapping the positions of training sample i and training sample k. σ is a constant, e is a constant, C is a loss function, λ ij is the first gradient, and the is used to indicate differentiation.

[0040] On the other hand, a computer device is provided. The computer device includes a processor and a memory. At least one computer program is stored in the memory and is loaded and executed by the processor so that the computer device implements the method for map search and sorting described in any one of the above.

[0041] On the other hand, a computer-readable storage medium is also provided. At least one computer program is stored in the computer-readable storage medium and is loaded and executed by a processor so that a computer implements the method for map search and sorting described in any one of the above.

[0042] On the other hand, a computer program product or a computer program is also provided. The computer program product or the computer program includes computer instructions, and the computer instructions are stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions so that the computer device executes the method for map search and sorting described in any one of the above.

[0043] The technical solutions provided in the embodiments of the present application at least bring the following beneficial effects:

[0044] By determining a visual position order that conforms to the map form, the search quality index obtained based on this visual position order is more in line with the actual situation during map search. On this basis, the map search and sorting model is trained with a search quality index that is more in line with the actual situation, and finally the obtained map search and sorting model has good performance. When performing map search and sorting based on this map search and sorting model, the sorting accuracy is high. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] To more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0046] Figure 1 is a schematic diagram of an implementation environment provided by an embodiment of the present application;

[0047] Figure 2 is a flowchart of a method for map search and sorting provided by an embodiment of the present application;

[0048] Figure 3 is a schematic diagram of a visual position order provided by an embodiment of the present application;

[0049] Figure 4It is a statistical chart of map location loss provided by an embodiment of the present application;

[0050] Figure 5 It is a schematic diagram of a device for map search sorting provided by an embodiment of the present application;

[0051] Figure 6 It is a schematic diagram of the structure of a server provided by an embodiment of the present application;

[0052] Figure 7 It is a schematic diagram of the structure of a terminal provided by an embodiment of the present application. Detailed implementation manners

[0053] To make the objectives, technical solutions and advantages of the present application clearer, the embodiments of the present application will be further described in detail below with reference to the accompanying drawings.

[0054] An embodiment of the present application provides a method for map search sorting. Please refer to Figure 1 , which shows a schematic diagram of the method implementation environment provided by an embodiment of the present application. The implementation environment may include: a terminal 11 and a server 12.

[0055] Among them, the terminal 11 is installed with an application program capable of obtaining a search request. When the application program obtains a search request, it sends the obtained search request to the server 12, and the server 12 can perform map search sorting based on the method provided by an embodiment of the present application. Optionally, the terminal 11 may obtain a map search sorting model from the server 12 and perform map search sorting based on the map search sorting model.

[0056] Alternatively, the terminal 11 is installed with an application program capable of obtaining a search request. When the application program obtains a search request, it can perform map search sorting based on the method provided by an embodiment of the present application. Optionally, the server 12 may obtain a map search sorting model from the terminal 11 and perform map search sorting based on the map search sorting model.

[0057] Alternatively, the terminal 11 obtains a search request from the server 12 and performs map search sorting based on the method provided by an embodiment of the present application. Optionally, the server 12 may obtain a map search sorting model from the terminal 11 and perform map search sorting based on the map search sorting model.

[0058] Optionally, the terminal 11 can be any kind of electronic product that can interact with the user through one or more means such as a keyboard, a touchpad, a touch screen, a remote control, voice interaction, or a handwriting device. For example, a PC (Personal Computer), a mobile phone, a smart phone, a PDA (Personal Digital Assistant), a wearable device, a PPC (Pocket PC), a tablet computer, a smart in-vehicle unit, a smart TV, a smart speaker, etc. The server 12 can be a single server, a server cluster composed of multiple servers, or a cloud computing service center. The terminal 11 and the server 12 establish a communication connection through a wired or wireless network.

[0059] Those skilled in the art should understand that the above-mentioned terminal 11 and server 12 are only examples. Other existing or future terminals or servers that can be applied to this application should also be included within the protection scope of this application and are hereby incorporated by reference.

[0060] Based on the above Figure 1 shown implementation environment, an embodiment of this application provides a method for map search sorting as Figure 2 shown. The method for map search sorting can be executed by the terminal or the server. Taking the application of this method to the terminal as an example, the method includes steps 201-step 203.

[0061] In step 201, a search request is obtained, and the visual position order of the map is determined based on the search request.

[0062] The embodiment of this application does not limit the manner of obtaining the search request. Optionally, an input box is provided, and the search request is input in the input box.

[0063] After the search request is obtained, the visual position order can be determined based on the search request. Exemplarily, determining the visual position order based on the search request includes: determining the starting point of the visual position order; determining the standard visual radius based on the starting point and the search request; determining the sorting strategy in the same visual distance interval, and determining the visual position order of the map based on the standard visual radius and the sorting strategy in the same visual distance interval.

[0064] Regarding the method for determining the starting point, optionally, the starting point is determined based on the center of the screen, the location of the user who initiates the search request, and the location of the sample to be sorted. For example, taking the screen as the plane where the rectangular coordinate system is located, setting the center of the screen as the origin of the rectangular coordinate system, and using the location of the sample to be sorted and the location of the user who initiates the search request as the coordinate values in the coordinate system. By performing vector summation operations in this rectangular coordinate system, the final dynamic visual center is obtained, and the calculated dynamic visual center is used as the starting point. Among them, the screen can be the screen of the device containing the map search sorting model, or the screen of the device that finally displays the sorting result of the map search sorting model. The embodiments of the present application do not limit this.

[0065] Exemplarily, the calculated dynamic visual center Center view has the following formula:

[0066] Center view =(X center , Y center ) (Formula 1)

[0067] Among them, x i is the coordinate value of the sample on the x-axis in the rectangular coordinate system. The sample includes the sample to be sorted and the user location in the screen view. α is the coordinate weight, i is the identifier of the sample index, and N is a positive integer; y i is the coordinate value of the sample on the y-axis in the rectangular coordinate system.

[0068] The embodiments of the present application do not limit the method for determining α. Exemplarily, the weight value is determined based on the sample type through an artificial strategy. Exemplarily, the weight value is determined based on the sample type through the method of other model learning. In a possible implementation, the weight α is defaulted to 1 through an artificial strategy. If the sample is a special sample, for example, when the sample is the user's location coordinate, the weight α is set to 2. When the sample is close to the subway station and the shopping mall, the weight α is set to 1.5.

[0069] Optionally, in addition to using the calculated dynamic visual center as the starting point, the screen visual center can also be used as the starting point. Among them, the screen can be the screen of the device containing the map search sorting model, or the screen of the device that finally displays the sorting result of the map search sorting model.

[0070] After determining the starting point, the standard visual radius can be determined based on the starting point Exemplarily, the formula for determining the standard visual radius is as follows:

[0071]

[0072] Among them, Discoordinate is the coordinate distance, representing the distance between the coordinates of the samples to be sorted on the screen and the starting point; View dis is the scale parameter. The embodiments of the present application do not limit the determination method of the scale parameter. Exemplarily, based on the area range of the search request, the scale parameter is determined to be 1:10,000.

[0073] After determining the standard visual radius, the visual distance intervals can be divided based on the standard visual radius and the screen size. Within the same visual distance interval, the sorting strategy is determined. The embodiments of the present application do not limit the determination method of the sorting strategy under the same visual distance interval. In a possible implementation, for different positions, according to the human visual order, the visual order is divided from left to right and from top to bottom for distinction. After determining the sorting strategy under the same visual distance interval, the visual position order can be determined based on the standard visual radius and the sorting strategy under the same visual distance interval.

[0074] Taking the division of the visual order from left to right and from top to bottom according to the human visual order as an example, the visual position order is finally determined, as Figure 3 shown: 1A, 1B, 1C, 1D, 2A, 2B…, and the expression is as follows:

[0075]

[0076] Formula 3 is the expression function of the visual position order where, is the standard visual distance radius.

[0077] In step 202, based on the visual position order, a search quality indicator that conforms to the map form is obtained, and the search quality indicator is used to indicate the accuracy of the search sorting.

[0078] The embodiments of the present application do not limit the method of obtaining the search quality indicator that conforms to the map form based on the visual position order. Optionally, a first data set and a second data set are obtained. The first data set is the click-through rate of the first sample in different views, and the second data set is the evaluation data of multiple second samples in the same view. The positions of the first sample and the multiple second samples are both within the reference range, and the reference range is determined based on the search request; based on the first data set, the second data set, and the visual position order, a target position loss function that conforms to the map form is determined, and the target position loss function is used to calculate the position loss in the map form; based on the target position loss function, a search quality indicator that conforms to the map form is determined.

[0079] The embodiments of the present application do not limit the manner of obtaining the first data set and the second data set, including but not limited to: logging in to the software background to obtain relevant data of the samples to be sorted within the scope of the search request. Taking the first data set as an example, logging in to the software background to obtain the click times and exposure times of the first sample, and calculating the click-through rate by calculating the click times and exposure times. Among them, the exposure times are the number of times the first sample is displayed on the view based on the search request. The calculated click-through rates at different positions are used as the first data set. Taking the second data set as an example, logging in to the software background to obtain the coordinates and click-through rates of multiple second samples to be sorted under the same view as the second data set.

[0080] After obtaining the first data set and the second data set, the target position loss function can be determined based on the first data set, the second data set, and the visual position order. Regarding the manner of determining the target position loss function, including but not limited to: determining the intermediate position loss function through the first data set and the visual position order; obtaining the target position loss function based on the intermediate position loss function and the second data set.

[0081] Regarding the manner of determining the intermediate position loss function based on the first data set and the visual position order, in a possible implementation manner, taking the distribution form of the position loss function in the form of a map and the position loss function in the form of a traditional list as an example, both conforming to the power-exponential family cluster distribution, the expression of the intermediate position loss function is obtained by fitting the data distribution of the first data set. Exemplarily, the expression of the intermediate position loss function is as follows:

[0082]

[0083] Among them, is the intermediate position loss function, α is the parameter for correcting the curvature of the loss function, β is the parameter for the amplitude of the loss function, is the visual position order.

[0084] After obtaining the intermediate position loss function, the target position loss function can be determined based on the second data set and the intermediate position loss function. By fitting the data distribution of the second data set, the values of α and β are solved, and the values of α and β are substituted into the intermediate position loss function to determine the target position loss function.

[0085] In a possible implementation manner, based on fitting the data distribution of the second data set, α = 0.035 and β = 0.25 are solved, and the values of α and β are substituted into the expression of the intermediate position loss function. The expression of the finally determined target position loss function is as follows:

[0086]

[0087] Formula 5 is the expression of the target position loss function for samples in the map form. By substituting the visual position order of the samples into Formula 5, the theoretical position loss of the samples in the map form can be obtained. The position loss statistics are as Figure 4 shown. Among them, the bar chart is the position loss statistically obtained in the map view, and the line chart is the theoretical position loss calculated based on Formula 5.

[0088] After determining the expression of the target position loss function, the search quality metric NDCG can be determined based on the expression of the target position loss function. DisView . In one possible implementation, the search quality metric is as follows:

[0089]

[0090] Among them, IDCG is the cumulative gain of loss in the ideal case. The ideal case is that the training samples are arranged in order according to their relevance to the search request. k represents the sorting position of the sample, G(k) represents the gain of the sample, and η(k) represents the position loss, with lower value as the position goes further back. y k then represents the relevance score of the sample. The higher the score, the greater the gain. Generally, y k is evaluated based on whether it is clicked. represents the visual position order.

[0091] In step 203, a map search ranking model is obtained based on the search quality metric, and map search ranking is performed through the map search ranking model.

[0092] The embodiments of the present application do not limit the manner of obtaining a map search ranking model based on the search quality metric and performing map search ranking through the map search ranking model. In one possible implementation, an expression of the first gradient is obtained based on the search quality metric, and the expression of the first gradient is used to determine the training direction and intensity; a map search ranking model is trained based on the expression of the first gradient.

[0093] Taking the initial map search ranking model to be trained as the traditional LambdaRank model (a search ranking model) as an example, the expression of the first gradient λ is obtained by multiplying the change amount of the search quality metric on the expression of the gradient of the traditional LambdaRank model. ij The expression of the first gradient is as follows:

[0094]

[0095] Among them, |Δ NDCG | is the change amount of the search quality metric after swapping the positions of training sample i and training sample k. S i represents the relevance score of training sample i. Sk The score representing the relevance of training sample k Used to indicate differentiation, C is the loss function, σ is a constant, and e is a constant. Optionally, e is 2.718281. Exemplarily, σ only affects the shape of the function graph and has little impact on the final result.

[0096] Using formula 7, the direction and intensity of adjusting the sample in the next iteration of the ranking model can be calculated. Update the parameters of the map search ranking model according to the training direction and intensity.

[0097] In a possible implementation, after obtaining the expression of the first gradient, use the initial map search ranking model to perform the first search ranking according to the search request, determine the direction and intensity of the next iteration based on the expression of the first gradient, and continuously adjust the ranking result through training. The embodiments of the present application do not limit the way of adjusting the ranking result. Optionally, reduce the visual impact of samples with low relevance to the search request by reducing the size of the displayed icons.

[0098] Optionally, after obtaining the map search ranking model that ends training, map search ranking can be performed based on this search ranking model. Input keywords, obtain samples related to the keywords, sort the samples based on the map search ranking model, and display the samples on the screen according to the ranking result.

[0099] In summary, the method for map search ranking provided by the embodiments of the present application determines a starting point for the samples to be ranked on the map, and the visual position order divided based on this starting point conforms to the sample distribution on the map. On this basis, the search quality index obtained according to the visual position order is more in line with the actual situation during map search. Train the initial map search ranking model through a search quality index that is more in line with the actual situation, update the parameters of the initial map search ranking model, and improve the performance of the trained map search ranking model. Therefore, when performing search ranking based on this map search ranking model, the accuracy of the search ranking result is higher.

[0100] In a possible implementation, the method for map search ranking provided by the embodiments of the present application can be applied to the offline state. Exemplarily, enter "food" in the input box as the search request, determine the visual position order of the map based on the store area that meets the food category; obtain the search quality index that conforms to the map form based on the visual position order; obtain the map search ranking model based on the search quality index, use the map search ranking model to perform map search ranking, and display the search results related to food on the device.

[0101] The above method can effectively improve the performance in the map search ranking task in the offline state, and the optimization effect is shown in Table 1:

[0102] Table 1

[0103]

[0104] As can be seen from Table 1, from the perspective of normalized discounted cumulative gain, a model evaluation metric, the solution of this application is compared with the reference algorithm. The reference algorithm is a map search ranking model trained based on existing search quality metrics. The map search ranking model trained by applying the technical solution provided in the embodiments of this application has better performance in the offline search ranking task.

[0105] See Figure 5 , the embodiments of this application provide a device for map search ranking. The device includes: a determination module 501, an acquisition module 502, and a ranking module 503.

[0106] The determination module 501 is configured to obtain a search request and determine the visual position order of the map based on the search request;

[0107] The acquisition module 502 is configured to obtain search quality metrics that conform to the map form based on the visual position order. The search quality metrics are used to indicate the accuracy of search ranking;

[0108] The ranking module 503 is configured to obtain a map search ranking model based on the search quality metrics and perform map search ranking through the map search ranking model.

[0109] Optionally, the determination module 501 is configured to determine the starting point of the visual position order; determine a standard visual radius based on the starting point and the search request; determine a ranking strategy in the same visual distance interval, and determine the visual position order of the map based on the standard visual radius and the ranking strategy in the same visual distance interval.

[0110] Optionally, the acquisition module 502 is configured to obtain a first data set and a second data set. The first data set is the click-through rate of a first sample under different views, and the second data set is the evaluation data of multiple second samples under the same view. The positions of the first sample and the multiple second samples are both within a reference range, and the reference range is determined based on the search request; determine a target position loss function that conforms to the map form based on the first data set, the second data set, and the visual position order. The target position loss function is used to calculate the position loss in the map form; determine search quality metrics that conform to the map form based on the target position loss function.

[0111] Optionally, the search quality metric is:

[0112]

[0113] Among them, IDCG is the cumulative gain loss in the ideal case. The ideal case is that the training samples are arranged in order according to their relevance to the search request. k represents the sorting position of the training sample, G(k) represents the gain of the training sample, η(k) represents the position loss of the training sample, and the position loss is determined based on the visual position order. NDCG DisView represents the search quality metric.

[0114] Optionally, the sorting module 503 is configured to obtain an expression of the first gradient based on the search quality metric, and the expression of the first gradient is used to determine the training direction and intensity; a map search sorting model is trained based on the expression of the first gradient.

[0115] Optionally, the expression of the first gradient is:

[0116]

[0117] where i and k represent the positions of the training samples, and S i represents the relevance score of training sample i, and S k represents the relevance score of training sample k, and Δ NDCG represents the change in the search quality metric after swapping the positions of training sample i and training sample k. σ is a constant, e is a constant, C is a loss function, and λ ij is the first gradient, used to indicate differentiation.

[0118] During the map search sorting process, the above device determines a visual position order that conforms to the map form, and the search quality metric obtained based on this visual position order is more in line with the actual situation during map search. On this basis, the map search sorting model is trained with a search quality metric that is more in line with the actual situation, and finally the obtained map search sorting model has good performance, and the sorting accuracy is high when performing map search sorting based on this map search sorting model.

[0119] It should be noted that when the above device provided in the embodiment realizes its functions, only the above-mentioned division of each functional module is used for illustration. In actual applications, the above functions can be allocated to different functional modules according to needs, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the device provided in the above embodiment and the method embodiment belong to the same concept, and the specific implementation process can be seen in the method embodiment, which will not be elaborated here.

[0120] Figure 6It is a schematic structural diagram of a server provided by an embodiment of the present application. The server may vary greatly due to different configurations or performances, and may include one or more processors (Central Processing Units, CPUs) 601 and one or more memories 602. Among them, at least one computer program is stored in the one or more memories 602, and the at least one computer program is loaded and executed by the one or more processors 601, so that the server implements the map search and sorting method provided by each of the above method embodiments. Of course, the server may also have components such as a wired or wireless network interface, a keyboard, and an input / output interface for input and output. The server may also include other components for implementing device functions, which will not be elaborated here.

[0121] Figure 7 It is a schematic structural diagram of a terminal provided by an embodiment of the present application. For example, the terminal is a smart phone, a tablet computer, an MP3 (Moving Picture Experts Group Audio Layer III) player, an MP4 (Moving Picture Experts Group Audio Layer IV) player, a laptop computer, or a desktop computer. The terminal may also be referred to by other names such as user equipment, portable terminal, laptop terminal, desktop terminal, etc.

[0122] Generally, the terminal includes: a processor 701 and a memory 702.

[0123] The processor 701 may include one or more processing cores, such as a quad-core processor, an octa-core processor, etc. The processor 701 may be implemented in at least one of the following hardware forms: DSP (Digital Signal Processing), FPGA (Field-Programmable Gate Array), and PLA (Programmable Logic Array). The processor 701 may also include a main processor and a coprocessor. The main processor is used to process data in the wake state and is also called the CPU (Central Processing Unit); the coprocessor is a low-power processor used to process data in the standby state. In some embodiments, the processor 701 may be integrated with a GPU (Graphics Processing Unit), and the GPU is responsible for rendering and drawing the content to be displayed on the display screen. In some embodiments, the processor 701 may further include an AI (Artificial Intelligence) processor, which is used to process computational operations related to machine learning.

[0124] The memory 702 may include one or more computer-readable storage media, and the computer-readable storage media may be non-transitory. The memory 702 may also include high-speed random access memory and non-volatile memory, such as one or more disk storage devices and flash storage devices. In some embodiments, the non-transitory computer-readable storage media in the memory 702 is used to store at least one instruction, and the at least one instruction is used to be executed by the processor 701 so that the terminal implements the map search and sorting method provided in the method embodiments of the present application.

[0125] In some embodiments, the terminal may optionally further include: a peripheral device interface 703 and at least one peripheral device. The processor 701, the memory 702, and the peripheral device interface 703 may be connected through a bus or signal lines. Each peripheral device may be connected to the peripheral device interface 703 through a bus, signal lines, or a circuit board. Specifically, the peripheral devices include at least one of the following: a radio frequency circuit 704, a display screen 705, a camera assembly 706, an audio circuit 707, a positioning component 708, and a power supply 709.

[0126] The peripheral device interface 703 can be used to connect at least one I / O (Input / Output) related peripheral device to the processor 701 and the memory 702. In some embodiments, the processor 701, the memory 702, and the peripheral device interface 703 are integrated on the same chip or circuit board; in some other embodiments, any one or two of the processor 701, the memory 702, and the peripheral device interface 703 can be implemented on a separate chip or circuit board, and this embodiment does not limit this.

[0127] The radio frequency circuit 704 is used to receive and transmit RF (Radio Frequency) signals, also known as electromagnetic signals. The radio frequency circuit 704 communicates with the communication network and other communication devices through electromagnetic signals. The radio frequency circuit 704 converts an electrical signal into an electromagnetic signal for transmission, or converts the received electromagnetic signal into an electrical signal. Optionally, the radio frequency circuit 704 includes: an antenna system, an RF transceiver, one or more amplifiers, a tuner, an oscillator, a digital signal processor, a codec chipset, a subscriber identity module card, and so on. The radio frequency circuit 704 can communicate with other terminals through at least one wireless communication protocol. The wireless communication protocol includes but is not limited to: metropolitan area network, generations of mobile communication networks (2G, 3G, 4G, and 5G), wireless local area network, and / or WiFi (Wireless Fidelity) network. In some embodiments, the radio frequency circuit 704 may further include a circuit related to NFC (Near Field Communication), and this application does not limit this.

[0128] The display screen 705 is used to display the UI (User Interface). The UI may include graphics, text, icons, videos, and any combination thereof. When the display screen 705 is a touch display screen, the display screen 705 also has the ability to collect touch signals on or above the surface of the display screen 705. The touch signals can be input as control signals to the processor 701 for processing. At this time, the display screen 705 can also be used to provide virtual buttons and / or virtual keyboards, also known as soft buttons and / or soft keyboards. In some embodiments, there can be one display screen 705, which is provided on the front panel of the terminal; in other embodiments, there can be at least two display screens 705, which are respectively provided on different surfaces of the terminal or in a folding design; in other embodiments, the display screen 705 can be a flexible display screen, which is provided on the curved surface or folding surface of the terminal. Even, the display screen 705 can also be set as an irregular non-rectangular shape, that is, an irregular-shaped screen. The display screen 705 can be prepared using materials such as LCD (Liquid Crystal Display) and OLED (Organic Light-Emitting Diode).

[0129] The camera module 706 is used to capture images or videos. Optionally, the camera module 706 includes a front camera and a rear camera. Generally, the front camera is provided on the front panel of the terminal, and the rear camera is provided on the back of the terminal. In some embodiments, there are at least two rear cameras, which are respectively any one of a main camera, a depth-of-field camera, a wide-angle camera, and a telephoto camera, so as to realize the function of background blurring by fusing the main camera and the depth-of-field camera, the function of panoramic shooting by fusing the main camera and the wide-angle camera, and the VR (Virtual Reality) shooting function or other fusion shooting functions. In some embodiments, the camera module 706 can also include a flash. The flash can be a single-color-temperature flash or a two-color-temperature flash. The two-color-temperature flash refers to the combination of a warm-light flash and a cold-light flash, which can be used for light compensation under different color temperatures.

[0130] The audio circuit 707 may include a microphone and a speaker. The microphone is used to collect sound waves of the user and the environment, and convert the sound waves into electrical signals for input to the processor 701 for processing, or input to the radio frequency circuit 704 to achieve voice communication. For the purpose of stereo collection or noise reduction, there may be multiple microphones, which are respectively arranged at different parts of the terminal. The microphone may also be an array microphone or an omnidirectional collection microphone. The speaker is used to convert the electrical signals from the processor 701 or the radio frequency circuit 704 into sound waves. The speaker may be a traditional thin film speaker or a piezoelectric ceramic speaker. When the speaker is a piezoelectric ceramic speaker, it can not only convert electrical signals into sound waves audible to humans, but also convert electrical signals into sound waves inaudible to humans for uses such as ranging. In some embodiments, the audio circuit 707 may further include a headphone jack.

[0131] The positioning component 708 is used to locate the current geographical location of the terminal to achieve navigation or LBS (Location Based Service). The positioning component 708 may be a positioning component based on the GPS (Global Positioning System) of the United States, the Beidou system of China, the GLONASS system of Russia, or the Galileo system of the European Union.

[0132] The power supply 709 is used to supply power to each component in the terminal. The power supply 709 may be alternating current, direct current, a disposable battery, or a rechargeable battery. When the power supply 709 includes a rechargeable battery, the rechargeable battery may support wired charging or wireless charging. The rechargeable battery may also be used to support fast charging technology.

[0133] In some embodiments, the terminal further includes one or more sensors 710. The one or more sensors 710 include but are not limited to: an acceleration sensor 711, a gyroscope sensor 712, a pressure sensor 713, a fingerprint sensor 714, an optical sensor 715, and a proximity sensor 716.

[0134] The acceleration sensor 711 can detect the magnitude of acceleration on the three coordinate axes of the coordinate system established with the terminal. For example, the acceleration sensor 711 can be used to detect the components of the gravitational acceleration on the three coordinate axes. The processor 701 can control the display screen 705 to display the user interface in a landscape view or a portrait view according to the gravitational acceleration signal collected by the acceleration sensor 711. The acceleration sensor 711 can also be used for collecting game or user movement data.

[0135] The gyroscope sensor 712 can detect the body direction and rotation angle of the terminal. The gyroscope sensor 712 can cooperate with the acceleration sensor 711 to collect the 3D actions of the user on the terminal. Based on the data collected by the gyroscope sensor 712, the processor 701 can implement the following functions: motion sensing (such as changing the UI according to the user's tilt operation), image stabilization during shooting, game control, and inertial navigation.

[0136] The pressure sensor 713 can be disposed on the side frame of the terminal and / or the lower layer of the display screen 705. When the pressure sensor 713 is disposed on the side frame of the terminal, it can detect the holding signal of the user on the terminal, and the processor 701 can perform left / right hand recognition or shortcut operations according to the holding signal collected by the pressure sensor 713. When the pressure sensor 713 is disposed on the lower layer of the display screen 705, the processor 701 can control the operable controls on the UI interface according to the pressure operation of the user on the display screen 705. The operable controls include at least one of button controls, scroll bar controls, icon controls, and menu controls.

[0137] The fingerprint sensor 714 is used to collect the fingerprints of the user. The processor 701 can identify the user's identity according to the fingerprints collected by the fingerprint sensor 714, or the fingerprint sensor 714 can identify the user's identity according to the collected fingerprints. When the identified user identity is a trusted identity, the processor 701 authorizes the user to perform relevant sensitive operations, and the sensitive operations include unlocking the screen, viewing encrypted information, downloading software, making payments, and changing settings, etc. The fingerprint sensor 714 can be disposed on the front, back, or side of the terminal. When there are physical buttons or a manufacturer's Logo (trademark) on the terminal, the fingerprint sensor 714 can be integrated with the physical buttons or the manufacturer's Logo.

[0138] The optical sensor 715 is used to collect the ambient light intensity. In one embodiment, the processor 701 can control the display brightness of the display screen 705 according to the ambient light intensity collected by the optical sensor 715. Specifically, when the ambient light intensity is high, the display brightness of the display screen 705 is increased; when the ambient light intensity is low, the display brightness of the display screen 705 is decreased. In another embodiment, the processor 701 can also dynamically adjust the shooting parameters of the camera module 706 according to the ambient light intensity collected by the optical sensor 715.

[0139] The proximity sensor 716, also known as the distance sensor, is usually disposed on the front panel of the terminal. The proximity sensor 716 is used to collect the distance between the user and the front of the terminal. In one embodiment, when the proximity sensor 716 detects that the distance between the user and the front of the terminal is gradually decreasing, the processor 701 controls the display screen 705 to switch from the lit state to the off state; when the proximity sensor 716 detects that the distance between the user and the front of the terminal is gradually increasing, the processor 701 controls the display screen 705 to switch from the off state to the lit state.

[0140] Those skilled in the art can understand that Figure 7 the structure shown in does not constitute a limitation on the terminal, and may include more or fewer components than shown in the figure, or combine some components, or adopt a different component layout.

[0141] In an exemplary embodiment, there is also provided a computer device, which includes a processor and a memory, and at least one computer program is stored in the memory. The at least one computer program is loaded and executed by one or more processors so that the computer device implements any of the above-mentioned map search and sorting methods.

[0142] In an exemplary embodiment, there is also provided a computer-readable storage medium, in which at least one computer program is stored. The at least one computer program is loaded and executed by the processor of the computer device so that the computer implements any of the above-mentioned map search and sorting methods.

[0143] In one possible implementation, the above-mentioned computer-readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a compact disc read-only memory (CD-ROM), magnetic tape, floppy disk, and optical data storage device, etc.

[0144] In an exemplary embodiment, there is also provided a computer program product or a computer program, which includes computer instructions stored in a computer-readable storage medium. The processor of the computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions so that the computer device executes any of the above-mentioned map search and sorting methods.

[0145] It should be understood that the "plurality" mentioned herein refers to two or more. "And / or" describes the association relationship of associated objects and indicates that there can be three relationships. For example, A and / or B can represent three situations: A exists alone, A and B exist simultaneously, and B exists alone. The character " / " generally represents an "or" relationship between the associated objects before and after.

[0146] The above are only exemplary embodiments of the present application and are not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method for map search and sorting, characterized in that, The method includes: Obtaining a search request and determining the visual position order of the map based on the search request; Obtaining a search quality metric that conforms to the map form based on the visual position order, where the search quality metric is used to indicate the accuracy of search sorting; Obtaining a map search sorting model based on the search quality metric and performing map search sorting through the map search sorting model; The determining the visual position order of the map based on the search request includes: Determining the starting point of the visual position order; Determining a standard visual radius based on the starting point and the search request; Determining a sorting strategy for the same visual distance interval, and determining the visual position order of the map based on the standard visual radius and the sorting strategy for the same visual distance interval; The obtaining a search quality metric that conforms to the map form based on the visual position order includes: Obtaining a first data set and a second data set, where the first data set is the click-through rate of a first sample under different views, and the second data set is the evaluation data of multiple second samples under the same view. The positions of the first sample and the multiple second samples are both within a reference range, and the reference range is determined based on the search request; Determining a target position loss function that conforms to the map form based on the first data set, the second data set, and the visual position order, where the target position loss function is used to calculate the position loss under the map form; Determining the search quality metric that conforms to the map form based on the target position loss function; The search quality metric is: Among them, the IDCG is the cumulative gain loss in the ideal case. The ideal case is that the training samples are arranged in order according to their relevance to the search request. k represents the sorting position of the training sample. G(k) represents the gain of the training sample. η(k) represents the position loss of the training sample. The position loss is determined based on the visual position order. The NDCG DisView represents the search quality metric; The obtaining a map search sorting model based on the search quality metric includes: Obtaining an expression of a first gradient based on the search quality metric, where the expression of the first gradient is used to determine the training direction and intensity; Training to obtain the map search sorting model based on the expression of the first gradient.

2. The method according to claim 1, characterized in that, The expression of the first gradient is: wherein, i and k represent the positions of the training samples, and S i represents the relevance score of training sample i, and S k represents the relevance score of training sample k, and Δ NDCG represents the change in the search quality metric after swapping the positions of training sample i and training sample k, σ is a constant, e is a constant, C is a loss function, and λ ij is the first gradient, and is used to indicate differentiation.

3. A device for map search and sorting, characterized in that, The device includes: A determination module for obtaining a search request and determining the visual position order of the map based on the search request; An acquisition module for obtaining a search quality metric that conforms to the map form based on the visual position order, where the search quality metric is used to indicate the accuracy of search sorting; A sorting module for obtaining a map search sorting model based on the search quality metric and performing map search sorting through the map search sorting model; The determination module for determining the starting point of the visual position order; determining a standard visual radius based on the starting point and the search request; determining a sorting strategy for the same visual distance interval, and determining the visual position order of the map based on the standard visual radius and the sorting strategy for the same visual distance interval; The obtaining module is configured to obtain a first data set and a second data set. The first data set is the click-through rate of a first sample under different views, and the second data set is the evaluation data of multiple second samples under the same view. The positions of the first sample and the multiple second samples are both within a reference range, and the reference range is determined based on the search request; determine a target position loss function that conforms to the map form based on the first data set, the second data set, and the visual position order, where the target position loss function is used to calculate the position loss in the map form; determine the search quality index that conforms to the map form based on the target position loss function, and the search quality index is: Among them, the IDCG is the cumulative gain loss in the ideal case. The ideal case is that the training samples are arranged in order according to their relevance to the search request. k represents the sorting position of the training sample. G(k) represents the gain of the training sample, and η(k) represents the position loss of the training sample. The position loss is determined based on the visual position order. The NDCG DisView represents the search quality metric. The sorting module is used to obtain an expression of the first gradient based on the search quality metric. The expression of the first gradient is used to determine the training direction and intensity. The map search sorting model is trained based on the expression of the first gradient.

4. The device according to claim 3, characterized in that, The expression of the first gradient is: wherein, i and k represent the positions of the training samples, and S i represents the relevance score of training sample i, and S k represents the relevance score of training sample k, and Δ NDCG represents the change in the search quality metric after swapping the positions of training sample i and training sample k, σ is a constant, e is a constant, C is a loss function, and λ ij is the first gradient, and is used to indicate differentiation.

5. A computer device, characterized in that, The computer device includes a processor and a memory. At least one computer program is stored in the memory, and the at least one computer program is loaded and executed by the processor so that the computer device implements the method for map search and sorting according to any one of claims 1 to 2.

6. A computer-readable storage medium, characterized in that, At least one computer program is stored in the computer-readable storage medium, and the at least one computer program is loaded and executed by a processor so that a computer implements the method for map search and sorting according to any one of claims 1 to 2.

7. A computer program product, characterized in that,The computer program product includes a computer program or instruction, and the computer program or instruction is executed by a processor so that a computer implements the method for map search and sorting according to any one of claims 1 to 2.

Images