Point recommendation method and device, electronic equipment and storage medium

Abstract

This invention discloses a point-of-interest recommendation method, apparatus, electronic device, and storage medium. The method includes: acquiring a user's current captured image information; comparing the user's captured image with multiple file information in a reference file set to determine the target file information corresponding to the user's captured image; determining a corresponding trip frequency matrix based on the target file information and the current point-of-interest information; the target file information includes captured image information of the user at multiple trip points, and the trip frequency matrix describes the number of times the user appears at each trip point at different times, with pre-calculated distance similarity and time similarity between multiple trip points; determining target interest points based on the trip frequency matrix, and recommending points of interest based on the target interest points. This makes the point-of-interest recommendation significantly more reasonable, capable of addressing user interest needs at different times and locations, and improving the accuracy and recall of the recommendation.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, specifically to a location recommendation method, apparatus, electronic device, and storage medium. Background Technology

[0002] Currently, when different users arrive at certain locations, points of interest can be recommended based on their historical records of visiting specific locations. However, existing recommendation methods are usually based on geographic location information or different time information, which makes point of interest recommendations inherently objective and unable to address the user's interest needs at different times and locations, resulting in insufficient recommendation accuracy. Summary of the Invention

[0003] Firstly, the main objective of this invention is to provide a location recommendation method, comprising:

[0004] Obtain the user's current captured image information; the current captured image information includes the user's captured image, the current time information corresponding to the user's captured image, and the current location information corresponding to the user's captured image;

[0005] The user-captured image is compared with multiple file information in the reference file set to determine the target file information corresponding to the user-captured image;

[0006] A corresponding trip frequency matrix is ​​determined based on the target profile information and the current location information; the target profile information includes captured image information of the user at multiple travel locations, and the trip frequency matrix is ​​used to describe the number of times the user appears at each travel location at different times, and the multiple travel locations include pre-calculated distance similarity and time similarity;

[0007] Based on the trip frequency matrix, target points of interest are determined, and points of interest are recommended accordingly.

[0008] Optionally, before acquiring the user's current captured image information, the following steps are included:

[0009] Based on the personnel image features corresponding to the images to be archived captured at each travel point, a similarity set between the images to be archived is determined.

[0010] The images to be archived are clustered based on the similarity set, and the images that meet the similarity threshold are archived to obtain the reference archive set.

[0011] Optionally, after clustering the images to be archived based on the similarity set and archiving the images that meet the similarity threshold to obtain the reference archive set, the process includes:

[0012] Based on the reference file set, the location information corresponding to each travel point is determined, and the distance between each pair of travel points is calculated;

[0013] The distance similarity between the two pairs of travel points is determined based on the distance between them.

[0014] Optionally, after clustering the images to be archived based on the similarity set and archiving the images that meet the similarity threshold to obtain the reference archive set, the method further includes:

[0015] The frequency of a user's appearance at a travel location in each time period is determined based on the aforementioned reference archive set;

[0016] A trip frequency matrix is ​​constructed based on the number of occurrences, and the trip frequency matrix is ​​normalized to obtain the matrix vector corresponding to the trip frequency matrix.

[0017] The temporal similarity between each pair of travel points is determined by calculation based on the matrix vector.

[0018] Optionally, determining the corresponding trip frequency matrix based on the target file information and the current location information includes:

[0019] Based on the target file information, determine the historical visit locations of the user;

[0020] Based on the current location information, determine the user's recommended location range;

[0021] Among the historical visited locations, those that fall within the recommended location range are designated as locations to be recommended.

[0022] Based on the captured image information of the user at the recommended location, the corresponding trip frequency matrix is ​​determined.

[0023] Optionally, determining the target points of interest based on the trip frequency matrix and making recommendations based on the target points of interest includes:

[0024] Based on the travel points corresponding to the travel frequency matrix, determine the time similarity and distance similarity of the travel points;

[0025] The total similarity is calculated based on the time similarity and distance similarity.

[0026] The target interest points are determined based on the total similarity, and recommendations are made based on the target interest points.

[0027] Optionally, determining the target interest point based on the total similarity and making recommendations based on the target interest point includes:

[0028] Based on the total similarity, each travel point is sorted to determine the target travel point whose total similarity meets the predetermined conditions;

[0029] The address of the point of interest corresponding to the target travel point is taken as the target point of interest, and recommendations are made based on the target point of interest.

[0030] Secondly, embodiments of the present invention provide a location recommendation device, comprising:

[0031] The acquisition module is used to acquire the user's current captured image information; the current captured image information includes the user's captured image, the current time information corresponding to the user's captured image, and the current location information corresponding to the user's captured image;

[0032] The comparison module is used to compare the user-captured image with multiple file information in the reference file set to determine the target file information corresponding to the user-captured image.

[0033] The determination module is used to determine the corresponding travel frequency matrix based on the target file information and the current location information; the target file information includes the captured image information of the user at multiple travel locations, and the travel frequency matrix is ​​used to describe the number of times the user appears at each travel location at different times, and the multiple travel locations include pre-calculated distance similarity and time similarity;

[0034] The recommendation module is used to determine the target interest points in the trip frequency matrix based on the current time information and the current location information, and to recommend points based on the target interest points.

[0035] Thirdly, embodiments of the present invention provide an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the point recommendation method described above.

[0036] Fourthly, embodiments of the present invention provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the point recommendation method described above.

[0037] The above-described solution of the present invention has at least the following beneficial effects:

[0038] The point-of-interest recommendation method provided by this invention first acquires the user's current captured image information. This information includes the captured image, the corresponding current time, and the corresponding current point of interest. The captured image is then compared with multiple files in a reference file set to determine the target file information corresponding to the captured image. A corresponding trip frequency matrix is ​​determined based on the target file information and the current point of interest. The target file information includes captured images of the user at multiple trip points, and the trip frequency matrix describes the number of times the user appears at each trip point at different times. Multiple trip points are compared using pre-calculated distance and time similarity. Based on the current time and current point of interest information, target points of interest are determined from the trip frequency matrix, and point-of-interest recommendations are made accordingly. This makes the point-of-interest recommendation highly reasonable, addressing user interest needs at different times and locations, and improving the accuracy and recall of the recommendations. Attached Figure Description

[0039] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the structures shown in these drawings without creative effort.

[0040] Figure 1 This is a schematic diagram of the overall process of the location recommendation method provided in the embodiments of the present invention;

[0041] Figure 2 This is a flowchart illustrating the location recommendation method provided in an embodiment of the present invention.

[0042] Figure 3 This is a structural block diagram of the location recommendation device provided in an embodiment of the present invention;

[0043] Figure 4 This is a structural block diagram of an electronic device provided in an embodiment of the present invention.

[0044] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0045] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0046] The terms "first," "second," and "third," etc., used in the specification, claims, and accompanying drawings of this invention are used to distinguish different objects and not to describe a specific order. Furthermore, the term "comprising," and any variations thereof, is intended to cover a non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or apparatuses.

[0047] First, let's take a look at the relevant accompanying drawings to illustrate the solution of the embodiments of this application.

[0048] like Figure 1 As shown, a specific embodiment of the present invention provides a location recommendation method, including:

[0049] S10. Obtain the user's current captured image information, which includes the user's captured image, the current time information corresponding to the user's captured image, and the current location information corresponding to the user's captured image.

[0050] In this embodiment, the current captured image information of the user can be obtained through the camera based on the user's recommendation request, or the current captured image information of the user can be obtained through the camera when the user arrives at a certain capture location. The camera can be set up in locations with high pedestrian traffic, such as traffic intersections, shopping malls, and streets, so as to collect user captured images in a timely manner. The user captured image can be a user's face image or a user's body image, etc. It is understood that the user's face image or body image can be image data that has been archived in history. For example, archiving images taken by the user within a month or 24 hours can determine the user's corresponding archived data. Therefore, the user's current captured image information can determine the user's corresponding archived data.

[0051] Specifically, before obtaining the user's current captured image information based on the user's recommendation request, the process includes: determining the similarity set between the images to be archived based on the personnel image features corresponding to the images to be archived captured at each travel point; clustering the images to be archived based on the similarity set, and archiving the images to be archived that meet the similarity threshold to obtain a reference archive set.

[0052] In this embodiment, the images to be archived captured at each travel point may include facial images or body images of different users. The similarity set can be calculated from the user's facial images. When acquiring user capture images at each travel point, the similarity of the user capture images can be calculated. For example, cosine similarity can be used to calculate the similarity between pairs of images to be archived. That is, the closer the cosine value between pairs of images to be archived is to 1, the greater the similarity between the pairs of archived data. It can be understood that after calculating the similarity set between the images to be archived, the data to be archived is clustered into corresponding image piles using the similarity set. Then, for each image pile, images with better image quality are determined as the initial archive set. Subsequently, user capture images collected at travel points can be archived according to the initial archive set, thereby obtaining a reference archive set. For example, in Figure 2 Each point shown can be represented as a file. After calculating the similarity of each file and clustering them, multiple image stacks can be formed, thereby completing the archiving operation between different files.

[0053] In a preferred embodiment, the reference archive set may be aidN{aid1, aid2, aid3...aidN}, and the captured images in each image archive set aidN can be represented as b. n {b1,b2,b3,..b n The similarity between pairs of captured images is calculated by performing n×n operations on the images captured at each travel point. Therefore, the calculated similarity set can be sim ij {sim 12 ,sim 13 ....sim ij}, i, j represent data b i b j After comparing the similarity set with the first similarity threshold, user-captured images with similarity scores greater than the first similarity threshold in the similarity set are retained. Therefore, the initial file set can be represented as aidN{aid1, aid2, aid3...aidN}. It can be understood that the first similarity threshold can be represented as α. sim When sim ij -α simWhen the similarity is greater than 0, the corresponding user-captured image is retained to form an initial archive set. When subsequent unarchived images collected at various travel points need to be archived, a similarity comparison is performed between the initial archive set and the unarchived images. Then, a second similarity threshold is applied. If the similarity is greater than the second similarity threshold, the user-captured image can be retained in the corresponding initial archive set to form a reference archive set. It can be understood that the second similarity threshold can be expressed as β. ij The similarity set determined by comparing subsequently acquired unarchived images with the initial archive set can be represented as sim. ij {sim 12 ,sim 13 ....sim ij}, thus in sim ij -β ij Images captured by users with a value greater than 0 can be archived in the corresponding archive data.

[0054] Furthermore, after clustering the images to be archived based on the similarity set and archiving the images that meet the similarity threshold to obtain the reference archive set, the process includes: determining the location information corresponding to each travel point based on the reference archive set and calculating the distance between each pair of travel points; and determining the distance similarity between each pair of travel points based on the distance between each pair of travel points.

[0055] In this embodiment, each user-captured image in the reference archive set includes corresponding travel point information and corresponding points of interest. The travel point information may include latitude and longitude, map coordinates, etc., and the points of interest may be scenic spots, schools, restaurants, etc., around the travel point. The distance similarity between each travel point is determined through the aforementioned reference archive set. The greater the distance between travel points, the smaller the distance similarity; the smaller the distance between travel points, the greater the distance similarity. Since users usually choose to visit points of interest that are closer to their location from each travel point, the distance similarity between each travel point can be inversely proportional to the distance between the user visiting each point of interest. The greater the distance, the smaller the distance similarity; the smaller the distance, the greater the distance similarity.

[0056] Understandably, the distance similarity mentioned above can be calculated using the following formula:

[0057]

[0058] distiance(li,lj)=R*arccos[sin(lat i )*sin(lat j)+cos(lat i )*cos(lat j )*cos(lat i -lon j )];

[0059] in, Distance(li,lj) represents the distance similarity between two travel points li and lj, and lat represents the distance between them. i and lon i The coordinates of the travel point are represented by latitude and longitude, and R is the Earth's radius: R = 6378.137 km. The distance between two travel points can be determined by the formula above. The distance between two travel points can be the road travel distance. The distance similarity between two travel points can be determined by the distance between them.

[0060] Furthermore, after clustering the images to be archived based on the similarity set and archiving the images that meet the similarity threshold to obtain the reference archive set, the process also includes: determining the number of times a user appears at a travel point in each time period based on the reference archive set; constructing a travel frequency matrix based on the number of occurrences and normalizing the travel frequency matrix to obtain a matrix vector; and calculating the temporal similarity between pairs of travel points based on the matrix vector.

[0061] In this embodiment, the trip frequency matrix can be a two-dimensional table matrix, and it is obtained by statistically analyzing the user's appearance at different travel points in different time periods. After normalizing the trip frequency matrix, multiple one-dimensional vector matrices are determined. These multiple one-dimensional vector matrices can be represented by matrix vectors, and then cosine similarity is calculated based on the matrix vectors, thereby determining the time similarity between pairs of travel points.

[0062] As is understandable, the table matrix corresponding to different time periods is shown below:

[0063]

[0064] Among them, users at various travel points carema n The table matrix showing the frequency of occurrences in different time periods is shown below:

[0065]

[0066]

[0067] As can be understood from the above table matrix, 24 hours can be divided into different time periods, and then the number of times multiple users appear at each travel point in each time period can be counted. For example, in the above table matrix, in the time period corresponding to t4, the number of times multiple users appear at the travel point carema1 is 56. This determines the user capture images collected by users in different time periods within 24 hours, and then counts the corresponding occurrences. After determining the above table matrix, normalization calculation is performed to determine the matrix vector.

[0068] Understandably, when performing normalization calculations using the aforementioned table matrix, the following formula can be used:

[0069]

[0070] in, Indicates travel location carema li The number of times each file appears at time point tj, N t Indicates travel location carema li The total number of occurrences, after calculating the one-dimensional vector matrix, can be represented as follows: Therefore, a matrix vector can be represented as After determining the matrix vectors, the temporal similarity between each pair of row points is calculated using cosine similarity, which can be performed using the following formula:

[0071] t_sim li,lj Indicates time similarity.

[0072] In calculating time similarity, the travel point corresponding to the current location information can be represented as carema. li Therefore, the current location information corresponds to the travel point carema li carema with various travel points n The temporal similarity between them can be calculated using the formula above. During the calculation, the travel time and number of trips corresponding to each travel point can be normalized using the mean and variance. That is, the travel time and number of trips in the above travel frequency matrix are normalized to a distribution with a mean of 0 and a variance of 1. Therefore, the resulting data has a mean of 0 and a variance of 1. Thus, the travel point information corresponding to the current location can be normalized. li First, the variance of the corresponding number of trips and trip times is calculated. Then, the carema is calculated based on the variance results and the normalization formula. li The corresponding normalized value; it can be understood that the travel point carema is obtained through calculation. liAfter normalization, multiple normalized values ​​can be used as travel point carema. li The matrix vectors are used to represent this, as described above. Therefore, after calculating the matrix vectors of all travel points, cosine similarity can be calculated for all travel point matrix vectors. This allows us to determine the temporal similarity between each pair of travel points. Temporal similarity can be expressed as the similarity between the travel time and the number of trips between each pair of travel points. Thus, by calculating the cosine similarity of the matrix vectors obtained from each pair of travel points, the temporal similarity between all travel points can be determined. In subsequent point-of-sale recommendations, when a user arrives at a certain travel point, corresponding points can be recommended based on the temporal similarity between that travel point and other travel points, improving the accuracy of point-of-sale recommendations. For example... In calculating time similarity, the travel points include A, B, C, and D. When a user appears at point A at 11:30 AM, point A can capture an image of the user at point A. Based on this image, the corresponding time information can be determined, and the user can be matched to determine that the user is in the time period t4. The corresponding point information can then be identified as point A. Therefore, when calculating time similarity, the number of trips corresponding to the time period t4 and point A can be normalized to obtain a normalized value. This normalized value can then be represented by a matrix vector to calculate the time similarity between point A and travel points B, C, and D in the time period t4.

[0073] S20. Compare the user-captured image with multiple file information in the reference file set to determine the target file information corresponding to the user-captured image.

[0074] In this embodiment, when a user appears at a certain travel point, the travel point can collect the user's current captured image information and perform similarity calculation with the aforementioned reference file set to determine the target file information in the reference file set. Therefore, the corresponding travel frequency matrix can be determined by searching through the target file information and the current captured image information, and then the corresponding recommended location can be determined.

[0075] S30. Determine the corresponding trip frequency matrix based on the target profile information and the current location information; the target profile information includes the captured image information of the user at multiple travel locations, and the trip frequency matrix is ​​used to describe the number of times the user appears at each travel location at different times. The multiple travel locations include pre-calculated distance similarity and time similarity.

[0076] In this embodiment, when the target file information is determined, the historical locations visited by the user can be found based on the target file information, and then the corresponding time similarity and distance similarity can be determined through the trip frequency matrix. It can be understood that the current captured image information includes the user captured image, the current time information corresponding to the user captured image, and the current location information corresponding to the user captured image. Therefore, the trip frequency matrix determined by the user within a historical time period can be obtained by determining the historical locations visited by the user.

[0077] Specifically, the above-mentioned determination of the corresponding trip frequency matrix based on target profile information and current location information includes: determining the user's historical visit locations based on target profile information; determining the user's location recommendation range based on current location information; selecting historical visit locations within the location recommendation range as potential recommendation locations; and determining the corresponding trip frequency matrix based on the captured image information of the user at the potential recommendation locations.

[0078] Here, "historical visited locations" refers to the locations a user has visited before the current time, and "recommended location range" represents the range of points of interest corresponding to the user's current location. This can be determined by calculating the similarity between the currently captured image information and each file in the reference file set. If the currently captured image information can identify a target file with a high similarity in the reference file set, then the user's historical visited locations can be determined based on the target file information. Optionally, each travel location has a corresponding recommended location range, which can be divided according to points of interest, such as schools, shopping malls, restaurants, hotels, and scenic spots. Therefore, points of interest can include learning, shopping, eating, accommodation, and sightseeing. The recommended location range is determined based on the current location information. This allows for the identification of historically visited locations within the recommended location range of the current location, which are then designated as potential recommended locations. Based on captured images of the user at these potential locations, a trip frequency matrix is ​​generated. This matrix determines the number of trips the user makes at each potential location, leading to more accurate recommendations. For example, if historically visited locations are A, B, C, and D, and the user's current location is A, and the recommended location range for A includes locations B and C, then locations B and C can be designated as potential recommended locations. The captured images of the user at locations B and C are then used to determine the corresponding number of trips, which are included in the trip frequency matrix for location recommendations.

[0079] S40. Determine the target points of interest based on the trip frequency matrix, and make recommendations based on the target points of interest.

[0080] In this embodiment, the target point of interest can be determined based on the time similarity and distance similarity mentioned above. Therefore, the corresponding travel point can be determined based on the travel frequency matrix. The distance between this travel point and the user's current travel point can be relatively close, and the user has arrived at this point many times in the past period. Therefore, the location of this travel point can be determined by comprehensively calculating the distance similarity and time similarity, and the associated target point of interest can be determined through this travel point.

[0081] In an optional embodiment, when determining the target point of interest location through distance similarity and time similarity, the distance similarity and time similarity can be filtered first to determine the distance similarity and time similarity with the maximum result. Then, a comprehensive calculation is performed based on the filtered distance similarity and time similarity to determine the corresponding travel point.

[0082] Specifically, the process of determining target points of interest based on the trip frequency matrix and making recommendations based on these target points of interest includes: determining the time similarity and distance similarity of the trip points based on the trip points corresponding to the trip frequency matrix; calculating the total similarity based on the time similarity and distance similarity; determining the target points of interest based on the total similarity; and making recommendations based on these target points of interest.

[0083] In this embodiment, the time similarity can be calculated using the aforementioned formula, and the distance similarity can be calculated using the aforementioned formula. After determining the trip frequency matrix based on the user's current captured image information, the trip location corresponding to the current captured image information can be determined. The distance similarity is determined based on this trip location and each other trip location. At the same time, the time similarity is determined based on the aforementioned trip frequency matrix. Then, the distance similarity and time similarity are weighted and summed to obtain the total similarity. The target interest point is determined through the total similarity. This approach comprehensively considers the user's interest needs corresponding to different times and locations, thereby improving the accuracy of interest point recommendation.

[0084] Understandably, the following formula can be used to calculate the total similarity mentioned above:

[0085] sim(li,lj)=α*t_sim li,lj +(1-α)d_sim li,lj ,

[0086] t_sim li,lj Time similarity;

[0087] d_sim li,lj Distance similarity

[0088] Where α∈[0,1] represents the weight value. t_sim represents the distance similarity between two travel points li and lj. li,lj The temporal similarity is calculated from the trip frequency matrix corresponding to the currently captured image information. sim(li,lj) represents the total similarity. The target interest point is determined by the total similarity, which allows for comprehensive recommendation by combining information from different time periods, resulting in higher recommendation accuracy.

[0089] Furthermore, target points of interest are determined based on the total similarity, and recommendations are made based on these target points of interest. This includes: sorting each travel point based on the total similarity to determine the target travel point whose total similarity meets the predetermined conditions; using the points of interest addresses corresponding to the target travel points as target points of interest, and making recommendations based on these target points of interest.

[0090] In this embodiment, after determining the total similarity, the trip points can be sorted in descending order of their total similarity. After sorting, the total similarity can be compared with a predetermined similarity. If the total similarity is greater than the predetermined similarity, the trip point corresponding to that total similarity can be used as the target trip point. If the total similarity is less than the predetermined similarity, the trip point corresponding to that total similarity can be eliminated. Therefore, when recommending to users, recommendations can be made based on the points of interest addresses corresponding to the target trip points, thereby improving the accuracy of point of interest recommendations and providing users with more diverse options.

[0091] In an optional embodiment, the travel points can be sorted in descending order of total similarity and the travel point with the highest total similarity can be determined as the target travel point. Thus, the point of interest address can be determined based on the target travel point with the highest total similarity and then recommended to the user, which can significantly improve the accuracy and recall rate of point of interest recommendation.

[0092] The point-of-interest recommendation method provided by this invention first acquires the user's current captured image information. This information includes the captured image, the corresponding current time, and the corresponding current point of interest. The captured image is then compared with multiple files in a reference file set to determine the target file information corresponding to the captured image. A corresponding trip frequency matrix is ​​determined based on the target file information and the current point of interest. The target file information includes captured images of the user at multiple trip points, and the trip frequency matrix describes the number of times the user appears at each trip point at different times. Multiple trip points are compared using pre-calculated distance and time similarity. Based on the current time and current point of interest information, target points of interest are determined from the trip frequency matrix, and point-of-interest recommendations are made accordingly. This makes the point-of-interest recommendation highly reasonable, addressing user interest needs at different times and locations, and improving the accuracy and recall of the recommendations.

[0093] It is understood that in the specific implementation of this application, data such as captured image information and file information are involved. When the embodiments in this application are applied to specific products or technologies, user permission or consent is required, and the collection, use and processing of related data must comply with the relevant laws, regulations and standards of the relevant countries and regions.

[0094] like Figure 3 As shown, this embodiment of the invention provides a location recommendation device 10, comprising:

[0095] The acquisition module 11 is used to acquire the user's current captured image information, which includes the user's captured image, the current time information corresponding to the user's captured image, and the current location information corresponding to the user's captured image.

[0096] The comparison module 12 is used to compare the user-captured image with multiple file information in the reference file set to determine the target file information corresponding to the user-captured image.

[0097] The determination module 13 is used to determine the corresponding trip frequency matrix based on the target file information and the current location information; the target file information includes the captured image information of the user at multiple travel locations, and the trip frequency matrix is ​​used to describe the number of times the user appears at each travel location at different times;

[0098] Recommendation module 14 is used to determine target points of interest based on the trip frequency matrix and to recommend points of interest based on the target points of interest.

[0099] The point-of-interest recommendation device 10 provided by this invention first acquires the user's current captured image information. This current captured image information includes the user's captured image, the current time information corresponding to the captured image, and the current point-of-interest information corresponding to the captured image. The device then compares the user's captured image with multiple file information in a reference file set to determine the target file information corresponding to the user's captured image. Based on the target file information and the current point-of-interest information, a corresponding trip frequency matrix is ​​determined. The target file information includes captured image information of the user at multiple trip points, and the trip frequency matrix describes the number of times the user appears at each trip point at different times. Multiple trip points include pre-calculated distance similarity and time similarity. Based on the current time information and the current point-of-interest information, target interest points are determined in the trip frequency matrix, and point-of-interest recommendations are made based on these target interest points. This makes the point-of-interest recommendation significantly more reasonable, capable of addressing user interest needs at different times and locations, and improving the accuracy and recall of the recommendations.

[0100] It should be noted that the location recommendation device 10 provided in the specific embodiment of the present invention is a device corresponding to the location recommendation method described above. All embodiments of the location recommendation method described above are applicable to the location recommendation device 10. Each embodiment of the location recommendation device 10 has a corresponding module corresponding to the steps in the location recommendation method described above, which can achieve the same or similar beneficial effects. In order to avoid excessive repetition, each module in the location recommendation device 2 will not be described in detail here.

[0101] like Figure 4 As shown, a specific embodiment of the present invention also provides an electronic device 20, including a memory 202, a processor 201, and a computer program stored in the memory 202 and executable on the processor 201. When the processor 201 executes the computer program, it implements the steps of the above-described point recommendation method.

[0102] Specifically, processor 201 calls the computer program stored in memory 202 and performs the following steps:

[0103] Obtain the user's current captured image information; the current captured image information includes the user's captured image, the current time information corresponding to the user's captured image, and the current location information corresponding to the user's captured image;

[0104] The user-captured image is compared with multiple file information in the reference file set to determine the target file information corresponding to the user-captured image;

[0105] The corresponding trip frequency matrix is ​​determined based on the target profile information and the current location information. The target profile information includes the captured image information of the user at multiple travel locations. The trip frequency matrix is ​​used to describe the number of times the user appears at each travel location at different times. The multiple travel locations include pre-calculated distance similarity and time similarity.

[0106] Based on the current time and location information, the target interest locations are determined in the trip frequency matrix, and location recommendations are made based on the target interest locations.

[0107] Optionally, before the processor 201 performs the process of acquiring the user's current captured image information, it includes:

[0108] Based on the characteristics of the people images captured at each travel point, a similarity set between the images to be archived is determined.

[0109] Clustering is performed on the images to be archived based on the similarity set, and the images to be archived that meet the similarity threshold are archived to obtain a reference archive set.

[0110] Optionally, after the processor 201 performs clustering of the images to be archived based on the similarity set and archives the images that meet the similarity threshold to obtain the reference archive set, it includes:

[0111] Based on the reference archive set, the location information corresponding to each travel point is determined, and the distance between each pair of travel points is calculated;

[0112] The similarity of distances between two travel points is determined based on the distance between each pair of travel points.

[0113] Optionally, after the processor 201 performs clustering of the images to be archived based on the similarity set and archives the images that meet the similarity threshold to obtain the reference archive set, it further includes:

[0114] The frequency of user visits to travel points in each time period is determined based on the reference archive set;

[0115] Construct a trip frequency matrix based on the number of occurrences, and then normalize the trip frequency matrix to obtain the matrix vector corresponding to the trip frequency matrix.

[0116] The temporal similarity between each pair of travel points is determined by calculation based on matrix vectors.

[0117] Optionally, the processor 201 executes a process to determine the corresponding trip frequency matrix based on the target file information and the current location information, including:

[0118] Based on the target file information, determine the historical visit locations of the user;

[0119] Based on the current location information, determine the recommended location range for the user;

[0120] Select historically visited locations that fall within the recommended location range as locations to be recommended.

[0121] Based on the captured image information of users at the locations to be recommended, a corresponding trip frequency matrix is ​​determined.

[0122] Optionally, the processor 201 performs the following steps: determining target points of interest based on the trip frequency matrix and making recommendations based on the target points of interest, including:

[0123] Based on the travel frequency matrix corresponding to the travel points, determine the time similarity and distance similarity of the travel points;

[0124] The total similarity is calculated based on time similarity and distance similarity.

[0125] The target interest points are determined based on the total similarity, and recommendations are made based on the target interest points.

[0126] Optionally, the processor 201 performs the following steps: determining the target interest point based on the total similarity, and making recommendations based on the target interest point, including:

[0127] The destinations are sorted according to the total similarity to determine the target destinations whose total similarity meets the predetermined conditions.

[0128] The address of the point of interest corresponding to the target travel point is used as the target point of interest, and recommendations are made based on the target point of interest.

[0129] That is, in a specific embodiment of the present invention, when the processor 201 of the electronic device 20 executes the computer program, it implements the steps of the above-mentioned point recommendation method, thereby making the point of interest recommendation obviously reasonable, able to solve the user interest needs corresponding to different times and locations, and improve the accuracy and recall rate of the recommendation.

[0130] It should be noted that since the processor 201 of the electronic device 20 implements the steps of the above-described location recommendation method when executing the computer program, all embodiments of the above-described location recommendation method are applicable to the electronic device 20 and can achieve the same or similar beneficial effects.

[0131] The computer-readable storage medium provided in this embodiment of the invention stores a computer program. When the computer program is executed by a processor, it implements the various processes of the point recommendation method or application endpoint recommendation method provided in this embodiment of the invention and can achieve the same technical effect. To avoid repetition, it will not be described again here.

[0132] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc.

[0133] In the description of this specification, references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0134] The above description is only a preferred embodiment of the present invention and does not limit the patent scope of the present invention. All equivalent structural transformations made under the concept of the present invention using the contents of the present invention specification and drawings, or direct / indirect applications in other related technical fields, are included within the patent protection scope of the present invention.

Claims

1. A method for recommending locations, characterized in that, include: Obtain the user's currently captured image information; The current captured image information includes the user captured image, the current time information corresponding to the user captured image, and the current location information corresponding to the user captured image; The user-captured image is compared with multiple file information in the reference file set to determine the target file information corresponding to the user-captured image; A corresponding trip frequency matrix is ​​determined based on the target profile information and the current location information; the target profile information includes captured image information of the user at multiple travel locations, and the trip frequency matrix is ​​used to describe the number of times the user appears at each travel location at different times, and the multiple travel locations include pre-calculated distance similarity and time similarity; Based on the current time information and the current location information, target interest locations are determined in the trip frequency matrix, and location recommendations are made based on the target interest locations; The step of determining the corresponding trip frequency matrix based on the target file information and the current location information includes: Based on the target file information, determine the historical visit locations of the user; Based on the current location information, the recommended location range for the user is determined, and the recommended location range is divided according to points of interest; Among the historical visited locations, those that fall within the recommended location range are designated as locations to be recommended. Based on the captured image information of the user at the recommended location, the corresponding trip frequency matrix is ​​determined.

2. The location recommendation method according to claim 1, characterized in that, Before obtaining the user's current captured image information, the following steps are included: Based on the personnel image features corresponding to the images to be archived captured at each travel point, a similarity set between the images to be archived is determined. The images to be archived are clustered based on the similarity set, and the images that meet the similarity threshold are archived to obtain the reference archive set.

3. The location recommendation method according to claim 2, characterized in that, The process of clustering the images to be archived based on the similarity set, and archiving the images that meet the similarity threshold to obtain the reference archive set, includes: Based on the reference file set, the location information corresponding to each travel point is determined, and the distance between each pair of travel points is calculated; The distance similarity between the two pairs of travel points is determined based on the distance between them.

4. The location recommendation method according to claim 2, characterized in that, After clustering the images to be archived based on the similarity set and archiving the images that meet the similarity threshold to obtain the reference archive set, the process further includes: The frequency of a user's appearance at a travel location in each time period is determined based on the aforementioned reference archive set; A trip frequency matrix is ​​constructed based on the number of occurrences, and the trip frequency matrix is ​​normalized to obtain the matrix vector corresponding to the trip frequency matrix. The temporal similarity between each pair of travel points is determined by calculation based on the matrix vector.

5. The location recommendation method according to claim 1, characterized in that, Based on the trip frequency matrix, target points of interest are determined, and recommendations are made based on these target points of interest, including: Based on the travel points corresponding to the travel frequency matrix, determine the time similarity and distance similarity of the travel points; The total similarity is calculated based on the time similarity and distance similarity. The target interest points are determined based on the total similarity, and recommendations are made based on the target interest points.

6. The location recommendation method according to claim 5, characterized in that, The step of determining the target interest point based on the total similarity and making recommendations based on the target interest point includes: Based on the total similarity, each travel point is sorted to determine the target travel point whose total similarity meets the predetermined conditions; The address of the point of interest corresponding to the target travel point is taken as the target point of interest, and recommendations are made based on the target point of interest.

7. A location recommendation device, characterized in that, include: The acquisition module is used to acquire the user's current captured image information; The current captured image information includes the user captured image, the current time information corresponding to the user captured image, and the current location information corresponding to the user captured image; The comparison module is used to compare the user-captured image with multiple file information in the reference file set to determine the target file information corresponding to the user-captured image. The determination module is used to determine the corresponding travel frequency matrix based on the target file information and the current location information; the target file information includes the captured image information of the user at multiple travel locations, and the travel frequency matrix is ​​used to describe the number of times the user appears at each travel location at different times, and the multiple travel locations include pre-calculated distance similarity and time similarity; The recommendation module is used to determine the target interest point in the trip frequency matrix based on the current time information and the current location information, and to recommend points based on the target interest point; The step of determining the corresponding trip frequency matrix based on the target file information and the current location information includes: Based on the target file information, determine the historical visit locations of the user; Based on the current location information, the recommended location range for the user is determined, and the recommended location range is divided according to points of interest; Among the historical visited locations, those that fall within the recommended location range are designated as locations to be recommended. Based on the captured image information of the user at the recommended location, the corresponding trip frequency matrix is ​​determined.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the location recommendation method as described in any one of claims 1 to 6.

9. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the location recommendation method as described in any one of claims 1 to 6.

Images