Invalid interest point detection method and device and electronic equipment

Abstract

The invention discloses an invalid interest point detection method and device and electronic equipment, and relates to the technical field of map making. The method comprises the following steps: acquiring a target area feature of map data of a target area type, thereby acquiring a target interest point density corresponding to the target area feature; wherein the target interest point density is calculated by a pre-trained regression model; acquiring map interest point density corresponding to the target area feature; and calculating a ratio of the map interest point density to the target interest point density, and if the ratio is greater than a ratio threshold, determining that invalid interest point information exists in the map data of the target area type. According to the embodiment, the technical problem that invalid interest point detection efficiency is low can be solved.

Description

Technical Field

[0001] This invention relates to the field of map production technology, and in particular to a method, apparatus, and electronic device for detecting invalid points of interest. Background Technology

[0002] To maintain the validity of Point of Interest (POI) information in map data, it is necessary to detect invalid POI information in the map data. Invalid POI information refers to POI information that exists in the map data but has actually disappeared.

[0003] Currently, invalid POI detection typically involves detecting invalidity on a single POI. For example, it might calculate the similarity between two POIs within a certain distance and then combine this with POI features to determine invalidity. Alternatively, it might calculate the behavior associated with the POI (such as WiFi (Wireless Fidelity) connection behavior, payment behavior, etc.) to determine invalidity. Therefore, existing methods for detecting invalid POIs target only individual POIs, resulting in low detection efficiency and a lack of ability to detect POIs lacking behavioral features. Summary of the Invention

[0004] In view of this, embodiments of the present invention provide a method, apparatus, and electronic device for detecting invalid points of interest, in order to solve the technical problem of low detection efficiency of invalid points of interest.

[0005] To achieve the above objectives, according to one aspect of the present invention, a method for detecting invalid points of interest is provided, comprising:

[0006] The target region features of the map data of the target region type are obtained, thereby obtaining the target interest point density corresponding to the target region features; wherein, the target interest point density is calculated by a pre-trained regression model;

[0007] Obtain the map interest point density corresponding to the features of the target area;

[0008] Calculate the ratio of the map interest point density to the target interest point density. If the ratio is greater than a ratio threshold, it is determined that there is invalid interest point information in the map data of the target area type.

[0009] Optionally, before obtaining the target interest point density corresponding to the target region features, the method further includes:

[0010] Obtain the sample region features and sample interest point density corresponding to the target region type;

[0011] The regression model is trained using the sample region features and the sample interest point density to obtain a trained regression model; wherein, the regression model is used to establish the mapping relationship between region features and interest point density;

[0012] The regression model is used to calculate the density of target interest points corresponding to the features of the target region.

[0013] Optionally, obtaining the map interest point density corresponding to the features of the target area includes:

[0014] Obtain the number of points of interest and the size of the target area's map data;

[0015] The number of interest points per unit size of the target region type is calculated based on the number of interest points and the region size, thereby obtaining the map interest point density corresponding to the target region feature.

[0016] Optionally, the target area features of the map data of the target area type are obtained, including:

[0017] Obtain the data to be detected for the target area in the map data;

[0018] The target region type of the target region is determined based on the data to be detected;

[0019] Obtain the target area features of the map data for the target area type.

[0020] Optionally, the target area features include at least one of the following: the city's tier, the city's region, and the ratio of the number of historical invalid points of interest to the number of current points of interest.

[0021] Optionally, if the target area feature is a region within the city, then the target area features of the map data of the target area type are obtained, including:

[0022] The coverage rate of building area within the first preset range of the target area type is determined based on the map data;

[0023] If the coverage rate is greater than the first coverage rate threshold, then the target area feature is determined to be the feature of the central area of ​​the city.

[0024] If the coverage rate is less than the first coverage rate threshold and greater than the second coverage rate threshold, then the target area is determined to be a combination of the central area and the suburbs of the city.

[0025] If the coverage rate is less than the second coverage rate threshold, then the target area feature is determined to be a feature of the suburban area of ​​the city.

[0026] Optionally, the target area type includes at least one of a first street, a second street, and a first building; wherein the number of points of interest belonging to the service type in the first street is greater than the number of points of interest belonging to the service type in the second street;

[0027] If the target area type is the first street or the second street, then the target area characteristics include at least one of the following: vehicle density on the street, number of parking lots connected to the street, and average number of floors of buildings on the street.

[0028] If the target area type is the first building, then the target area features include at least one of the following: the number of floors in the first building, the number of points of interest in the first building, and the number of points of interest within a second preset range of the first building.

[0029] Furthermore, according to another aspect of the present invention, an invalid point of interest detection device is provided, comprising:

[0030] The first acquisition module is used to acquire the target area features of map data of the target area type, thereby acquiring the target interest point density corresponding to the target area features; wherein, the target interest point density is calculated by a pre-trained regression model;

[0031] The second acquisition module is used to acquire the map interest point density corresponding to the features of the target area;

[0032] The detection module is used to calculate the ratio of the density of map points of interest to the density of target points of interest. If the ratio is greater than a ratio threshold, it is determined that there is invalid point of interest information in the map data of the target area type.

[0033] Optionally, a training module is also included for:

[0034] Obtain the sample region features and sample interest point density corresponding to the target region type;

[0035] The regression model is trained using the sample region features and the sample interest point density to obtain a trained regression model; wherein, the regression model is used to establish the mapping relationship between region features and interest point density;

[0036] The regression model is used to calculate the density of target interest points corresponding to the features of the target region.

[0037] Optionally, the second acquisition module is further configured to:

[0038] Obtain the number of points of interest and the size of the target area's map data;

[0039] The number of interest points per unit size of the target region type is calculated based on the number of interest points and the region size, thereby obtaining the map interest point density corresponding to the target region feature.

[0040] Optionally, the first acquisition module is further configured to:

[0041] Obtain the data to be detected for the target area in the map data;

[0042] The target region type of the target region is determined based on the data to be detected;

[0043] Obtain the target area features of the map data for the target area type.

[0044] Optionally, the target area features include at least one of the following: the city's tier, the city's region, and the ratio of the number of historical invalid points of interest to the number of current points of interest.

[0045] Optionally, if the target area feature is a region of the city where it is located, the first acquisition module is further configured to:

[0046] The coverage rate of building area within the first preset range of the target area type is determined based on the map data;

[0047] If the coverage rate is greater than the first coverage rate threshold, then the target area feature is determined to be the feature of the central area of ​​the city.

[0048] If the coverage rate is less than the first coverage rate threshold and greater than the second coverage rate threshold, then the target area is determined to be a combination of the central area and the suburbs of the city.

[0049] If the coverage rate is less than the second coverage rate threshold, then the target area feature is determined to be a feature of the suburban area of ​​the city.

[0050] Optionally, the target area type includes at least one of a first street, a second street, and a first building; wherein the number of points of interest belonging to the service type in the first street is greater than the number of points of interest belonging to the service type in the second street;

[0051] If the target area type is the first street or the second street, then the target area characteristics include at least one of the following: vehicle density on the street, number of parking lots connected to the street, and average number of floors of buildings on the street.

[0052] If the target area type is the first building, then the target area features include at least one of the following: the number of floors in the first building, the number of points of interest in the first building, and the number of points of interest within a second preset range of the first building.

[0053] According to another aspect of the present invention, an electronic device is also provided, comprising:

[0054] One or more processors;

[0055] Storage device for storing one or more programs.

[0056] When the one or more programs are executed by the one or more processors, the one or more processors implement the method described in any of the above embodiments.

[0057] According to another aspect of the present invention, a computer-readable medium is also provided, on which a computer program is stored, which, when executed by a processor, implements the methods described in any of the above embodiments.

[0058] According to another aspect of the present invention, a computer program product is also provided, including a computer program that, when executed by a processor, implements the methods described in any of the above embodiments.

[0059] One embodiment of the above invention has the following advantages or beneficial effects: Because it employs a technique of first acquiring the target region features of the map data of the target region type, thereby obtaining the target interest point density corresponding to the target region features, then acquiring the map interest point density corresponding to the target region features, and finally calculating the ratio of the map interest point density to the target interest point density, if the ratio is greater than a ratio threshold, it determines that invalid interest point information exists in the map data of the target region type. Therefore, it overcomes the technical problem of low efficiency in detecting invalid interest points in the prior art. This embodiment of the invention, by determining whether the ratio of the map interest point density to the target interest point density is greater than a ratio threshold, achieves the discovery of areas containing a large amount of invalid interest point information, that is, it achieves batch location of invalid interest point information, thereby improving the detection efficiency of invalid interest points.

[0060] The further effects of the aforementioned unconventional alternative methods will be explained below in conjunction with specific implementation methods. Attached Figure Description

[0061] 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 these drawings without creative effort.

[0062] Figure 1 This is a flowchart of an invalid interest point detection method according to a first embodiment of the present invention.

[0063] Figure 2 This is a flowchart of an invalid interest point detection method according to a second embodiment of the present invention.

[0064] Figure 3 This is a flowchart of an invalid interest point detection method according to a third embodiment of the present invention.

[0065] Figure 4 This is a schematic diagram of an invalid point of interest detection device according to an embodiment of the present invention.

[0066] Figure 5 This is an exemplary system architecture diagram in which embodiments of the present invention can be applied.

[0067] Figure 6 This is a schematic diagram of the structure of a computer system suitable for implementing terminal devices or servers of the present invention. Detailed Implementation

[0068] The following description, in conjunction with the accompanying drawings, illustrates exemplary embodiments of the present invention, including various details to aid understanding. These details should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the invention. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0069] It should be noted that the collection, gathering, updating, analysis, processing, use, transmission, and storage of user personal information involved in this disclosed technical solution all comply with relevant laws and regulations, are used for legitimate purposes, and do not violate public order and good morals. Necessary measures are taken to prevent unauthorized access to user personal information data and to safeguard user personal information security, network security, and national security.

[0070] Figure 1 This is a flowchart of an invalid interest point detection method according to a first embodiment of the present invention. As one embodiment of the present invention, such as... Figure 1 As shown, the method for detecting invalid points of interest may include:

[0071] Step 101: Obtain the target region features of the map data of the target region type, and thereby obtain the target interest point density corresponding to the target region features; wherein, the target interest point density is calculated by a pre-trained regression model.

[0072] In embodiments of the present invention, the target area type can be a large building, a commercial street, or an ordinary street, etc. A large building generally refers to a building whose Area of ​​Interest (AOI) area is larger than a preset area (e.g., 1000 square meters, 1200 square meters, or 1500 square meters), and the number of points of interest (POIs) within the AOI is greater than a first threshold (e.g., 10, 12, or 15), with at least half of the POIs being of service type (service types may include catering, retail, commerce, residential services, etc.). The remaining buildings can be considered non-large buildings. It should be noted that for non-large buildings, due to their smaller AOIs and fewer POIs, they lack reference value and are not considered within the scope of this embodiment. A commercial street generally refers to a street where the number of POIs within an AOI is greater than a second threshold (e.g., 20, 25, or 30), and at least half of the POIs are of service type (service types may include catering, retail, commerce, residential services, etc.). The second threshold and the first threshold can be the same or different. The remaining streets are considered ordinary streets. A street can refer to a complete street or a segment of a street. For example, a street can be divided into multiple road segments according to a preset length (e.g., 500 meters, 600 meters, or 800 meters). If a road segment meets the above conditions, it is a commercial street; otherwise, it is an ordinary street. Road segments that do not meet the preset length (e.g., a 100-meter, 150-meter, or 180-meter segment) can also be considered a complete road segment and processed according to the preset length. Map data can be pre-stored in a database for easy retrieval and to improve the efficiency of detecting invalid points of interest.

[0073] Optionally, the target area features include at least one of the following: the city's tier, the city's region, and the ratio of the historical invalidity of points of interest to the current number of points of interest. In embodiments of the present invention, the city's tier may include first-tier cities, second-tier cities, third-tier cities, and county towns, etc., and the city's tier can be obtained from the administrative divisions of points of interest recorded in the map data of the target area type. The city's region may include the central area, suburbs, and a combination of the central area and suburbs, etc. The historical invalidity of points of interest may refer to the number of invalid points of interest recorded within a preset time period up to the current time. For example, if the preset time period is 2 years, it means recording the number of invalid points of interest in the past 2 years, such as 50 invalid points of interest in the past 2 years. Assuming the current number of points of interest is 2000, then the ratio of the historical invalidity of points of interest to the current number of points of interest is 50 / 2000 = 1 / 40, which can also be regarded as the update rate of points of interest.

[0074] Optionally, if the target area feature is a region within the city, then obtaining the target area feature from the map data of the target area type includes: determining the coverage rate of the building area within a first preset range of the target area type based on the map data; if the coverage rate is greater than a first coverage rate threshold, then determining that the target area feature is a feature of the central area of ​​the city; if the coverage rate is less than the first coverage rate threshold but greater than a second coverage rate threshold, then determining that the target area feature is a feature of a combined central and suburban area of ​​the city; if the coverage rate is less than the second coverage rate threshold, then determining that the target area feature is a feature of the suburban area of ​​the city. In embodiments of the present invention, the coverage rate is the ratio of the building area within the first preset range to the area of ​​the first preset range. The first coverage rate threshold can be any value between 60% and 40%, such as 55%, 50%, 48%, or 42%, and the second coverage rate threshold can be any value between 30% and 20%, such as 28%, 25%, 23%, or 21%. Dividing the area according to the ratio of the building area to the area of ​​the first preset range yields a more accurate division result.

[0075] Optionally, the target area type includes at least one of a first street, a second street, and a first building; wherein the number of points of interest (POIs) belonging to the service type in the first street is greater than the number of POIs belonging to the service type in the second street. Optionally, if the target area type is the first street or the second street, the target area features include at least one of the following: vehicle density on the street, number of parking lots connected to the street, and average number of floors in the buildings on the street; if the target area type is the first building, the target area features include at least one of the following: number of floors in the first building, number of POIs in the first building, and number of POIs within a second preset range of the first building.

[0076] In embodiments of the present invention, the number of floors in a building can be obtained from map data stored in a database. The density of vehicles on a street can be the number of vehicles per unit length (e.g., 100 meters) of the street. The number of vehicles on a street can be obtained through image acquisition, thus yielding the number of vehicles per unit length. A parking lot connected to a street can refer to a parking lot where the street is directly connected to the exit or entrance of the parking lot. The average number of floors in buildings on the street (which can be the first building and the second building) is, for example, two 10-story buildings and two 5-story buildings, with an average of 7.5 floors. The number of points of interest (POIs) within a second preset range of the first building generally does not include the number of POIs within the first building. For example, if the first building has 100 POIs, and the preset range of the first building includes 200 shops (excluding the first building), and these 200 shops have 200 POIs, then the first building and the 200 shops together have 300 POIs, while the second preset range of the first building contains only 200 POIs. The second preset range can be preset, for example, any value within the range of 100-500 meters. The number of points of interest (POIs) within the first building can also be obtained through image acquisition. For example, by acquiring an image of the exterior of the first building (where "exterior" can be understood as the outer wall), POIs within the image can be identified through text recognition. When the target area type is the first building, the target area features can also include the area of ​​the building's interest surface. When the target area type is the first street or the second street, the target area features can also include the number of POIs within the street. The number of POIs within the street can also be obtained using image recognition.

[0077] In embodiments of the present invention, the density of target interest points corresponding to the features of the target region can be pre-calculated and stored in a database, which helps to improve the detection efficiency of invalid interest points.

[0078] Optionally, before step 101, the method further includes: obtaining sample region features and sample interest point density corresponding to the target region type; training a regression model using the sample region features and sample interest point density to obtain a trained regression model; wherein, the regression model is used to establish a mapping relationship between region features and interest point density; and calculating the target interest point density corresponding to the target region features using the regression model. Before step 101, the regression model needs to be trained first. Specifically, each sample region feature corresponding to the target region type and its corresponding sample interest point density are obtained, and the regression model is trained using each sample region feature and its corresponding sample interest point density. Then, the target interest point density corresponding to the target region features is calculated using the trained regression model, and the calculation result is stored in the database. Optionally, there can be various regression models, such as linear regression, multinomial regression, or partial least squares regression. In this embodiment of the invention, linear regression model can be used preferentially for training. It should be noted that the target interest point density corresponding to the target region features calculated using the trained regression model can be regarded as the density of effective interest points.

[0079] Step 102: Obtain the map interest point density corresponding to the features of the target area.

[0080] Generally, the density of points of interest (POIs) in map data differs depending on the characteristics of the target region. Since the target region features are characteristics of the target region type, the density of POIs corresponding to these features is also the density of POIs for that target region type.

[0081] Step 103: Calculate the ratio of the map interest point density to the target interest point density. If the ratio is greater than the ratio threshold, it is determined that there is invalid interest point information in the map data of the target area type.

[0082] Considering the possibility of invalid points of interest (POIs) in the map data of the target area type, the density of map POIs corresponding to the target area features may be greater than the density of target POIs. The ratio threshold can be preset, for example, set to any value between 1.5 and 4, such as 1.5, 2, 2.8, 3, or 4. When the ratio of the map POI density to the target POI density is greater than the ratio threshold, it indicates the presence of multiple invalid POIs in the map data of the target area type.

[0083] Based on the various embodiments described above, it can be seen that the embodiments of the present invention first obtain the target area features of the map data of the target area type, thereby obtaining the target interest point density corresponding to the target area features, then obtaining the map interest point density corresponding to the target area features, and finally calculating the ratio of the map interest point density to the target interest point density. If the ratio is greater than a ratio threshold, it is determined that there is invalid interest point information in the map data of the target area type. This technical means solves the technical problem of low detection efficiency of invalid interest points in the prior art. By judging whether the ratio of the map interest point density to the target interest point density is greater than the ratio threshold, the embodiments of the present invention realize the discovery of areas where a large number of invalid interest point information is located, that is, realize the batch location of invalid interest point information, thereby improving the detection efficiency of invalid interest points.

[0084] Figure 2 This is a flowchart of an invalid interest point detection method according to a second embodiment of the present invention. As another embodiment of the present invention, such as... Figure 2 As shown, the method for detecting invalid points of interest may include:

[0085] Step 201: Obtain the sample region features and sample interest point density corresponding to the target region type.

[0086] Step 202: The regression model is trained using the sample region features and the sample interest point density to obtain a trained regression model; wherein, the regression model is used to establish the mapping relationship between region features and interest point density.

[0087] Step 203: Obtain the target area features of the map data of the target area type, and use the regression model to calculate the target interest point density corresponding to the target area features.

[0088] In embodiments of the present invention, the sample interest point density corresponding to the target region type is a manually verified interest point density, meaning that the interest points involved in the calculation of the sample interest point density are valid interest points. For example, if the interest points in the target region type (such as large buildings, commercial streets, or ordinary streets) have been manually verified, they are considered valid interest points, and the sample interest point density of the target region type obtained using these valid interest points is also verified. The interest points included in the target region type are also one of the sample region features corresponding to the target region type. That is to say, the sample region features corresponding to the target region type can generally also be manually verified. However, the other sample region features corresponding to the target region type (sample region features other than the interest points included in the target region type) can be either manually verified or not. In this embodiment, the case of not being verified is preferred. The unverified sample region features (sample region features other than the interest points included in the target region type) can be kept in the same state as the corresponding target region features (both are unverified), so that the difference between the obtained map interest point density and the target interest point density is caused by invalid interest points in the interest points included in the target region type, making the detection of invalid interest points in the target region type more accurate.

[0089] In embodiments of the present invention, a regression model is trained using sample region features and sample interest point density to obtain a trained regression model. This trained regression model establishes a mapping relationship between region features and interest point density. Target region features can be substituted into the regression model, and the target interest point density corresponding to the target region features is calculated based on the established mapping relationship. Compared to existing solutions, embodiments of the present invention can detect regions with a large number of invalid interest points, making it more efficient and convenient, and helping to improve the detection efficiency of invalid interest points. Furthermore, a linear regression model is preferred. Optionally, the mapping relationship between region features and interest point density established by the linear regression model is shown in the formula:

[0090] h θ (x)=θ T x = θ0x0 + θ1x1 + ... + θ n x n

[0091] Among them, h θ (x) represents the interest point density value, x represents the region feature, and θ represents the coefficient of the region feature obtained after training.

[0092] In some other embodiments of the present invention, after obtaining the sample region features and sample interest point density corresponding to the target region type, the method may further include: normalizing the sample region features to obtain normalized sample features; and normalizing the target region features to obtain normalized target features. Optionally, training the regression model using the sample region features and sample interest point density may include: training the regression model using normalized sample features and sample interest point density. Optionally, calculating the target interest point density corresponding to the target region features using a pre-trained regression model may include: calculating the target interest point density corresponding to the normalized target features using the regression model trained with normalized sample features and sample interest point density.

[0093] In an embodiment of the present invention, the normalization function is as follows:

[0094]

[0095] Where x represents the feature before normalization, i.e., the initial feature value, z represents the feature after normalization, μ represents the mean, and σ represents the variance. μ and σ can be calculated based on the features of the sample region.

[0096] In this embodiment, the features of the sample region are normalized to obtain normalized sample features, and the features of the target region are normalized to obtain normalized target features. The dimensions of the features can be ignored, and the values ​​of each feature can be proportionally converted to a certain range, such as values ​​between -1 and 1, or any value between 0 and 1. Normalizing the features makes feature processing more convenient. Using normalized sample features and sample interest point density to train the regression model also helps improve the calculation accuracy of the trained regression model, thereby contributing to improved accuracy in detecting invalid interest points.

[0097] Step 204: Obtain the target interest point density corresponding to the target region features.

[0098] Step 205: Obtain the map interest point density corresponding to the features of the target area.

[0099] Step 206: Calculate the ratio of the map interest point density to the target interest point density.

[0100] Step 207: Determine whether the ratio is greater than the ratio threshold. If yes, proceed to step 208; otherwise, end the process.

[0101] Step 208: Determine if invalid points of interest exist in the map data of the target area type.

[0102] Therefore, by determining whether the ratio of map interest point density to target interest point density is greater than a ratio threshold, this embodiment of the invention can quickly locate areas with invalid interest point information, thus achieving batch location of invalid interest point information and improving the detection efficiency of invalid interest points.

[0103] Figure 3 This is a flowchart of an invalid interest point detection method according to a third embodiment of the present invention. As another embodiment of the present invention, such as... Figure 3 As shown, the method for detecting invalid points of interest may include:

[0104] Step 301: Obtain the data to be detected for the target area in the map data.

[0105] Step 302: Determine the target region type of the target region based on the data to be detected.

[0106] In embodiments of the present invention, the target area can be a building or a street. When the target area is a building, determining the target area type based on the data to be detected may include determining the area of ​​the building's interest surface based on the data to be detected. If the area is greater than a preset area, and the number of interest points in the interest surface is greater than a third quantity threshold, and at least half of the interest points are of service type, then the building is determined to be a large building (also referred to as a first building), and the rest are non-large buildings (also referred to as second buildings). When the target area is a street, determining the target area type based on the data to be detected may include determining the number of interest points contained in an interest surface of a street based on the data to be detected. If this number is greater than a fourth quantity threshold, and at least half of the interest points are of service type, then the street is determined to be a commercial street (also referred to as a first street), and the rest are ordinary streets (also referred to as second streets). Considering the significant differences in the number of points of interest (POIs) across different target areas (such as buildings and streets), and the substantial differences in POIs across different target area types within the same target area (such as commercial streets and ordinary streets), multiple target area types were defined to obtain the target area characteristics of the map data for each target area type. This enabled the calculation of POI density from multiple dimensions, which is beneficial for obtaining more accurate POI density based on different target area characteristics.

[0107] Step 303: Obtain the target area features of the map data of the target area type.

[0108] The target area type may include at least one of the following: first street, second street, and first building; wherein the number of points of interest belonging to the service type in the first street is greater than the number of points of interest belonging to the service type in the second street.

[0109] If the target area type is a first building (large building), then the target area characteristics include at least one of the following: the city level, which can include first-tier cities, second-tier cities, third-tier cities, and county towns; the city area, which can include the city center, the area combining the city center and suburbs, and the suburban area; the number of floors in the first building, which can be obtained from map data or through image recognition; the area of ​​the first building, which can be obtained from map data, i.e., the AOI area; the number of points of interest in the first building, which can be obtained through the image acquisition methods described above; the number of points of interest within a second preset range of the first building, which can be obtained from map data; and the ratio of the historical invalid number of points of interest to the current number of points of interest in the first building (the first building point of interest update rate), which can be obtained from map data by calculating the ratio of the historical invalid number to the current number of points of interest.

[0110] In embodiments of the present invention, the number of points of interest contained in different target region types is generally different. In order to perform targeted detection of invalid points of interest for different target region types and improve the accuracy of invalid point of interest detection, embodiments of the present invention divide regions that may contain points of interest into multiple different target region types.

[0111] It is understandable that each target area type can have a corresponding sample target area type, and each target area feature can have a corresponding sample area feature. For example, the city level has a corresponding sample city level, the city area has a corresponding sample city area, the number of floors in the first building has a corresponding sample first building number of floors, the area of ​​the first building has a corresponding sample first building area, the number of points of interest in the first building has a corresponding number of points of interest in the first building, the number of points of interest in the second preset range of the first building has a corresponding number of points of interest in the second preset range of the first building, and the point of interest update rate of the first building has a corresponding sample first building point of interest update rate.

[0112] It should be noted that the number of interest points corresponding to different target region features is generally different. When calculating the number and density of interest points, the more types of target region features used, the more accurate the calculation results will be. Therefore, the regression model can be trained using multiple different types of sample region features, so that the trained regression model can establish a mapping relationship between the interest point density in the target region type based on multiple different sample region features. This enables the determination of the interest point density in the target region type from multiple different dimensions of target region features, which has high accuracy and helps to improve the detection accuracy of invalid interest points.

[0113] When the target area type is First Street or Second Street, the target area characteristics include at least one of the following: the city level, which can include first-tier cities, second-tier cities, third-tier cities, and county towns; the city area, which can include the city center, the area combining the city center and suburbs, and the suburban area; the average number of floors of buildings on the street, which can be obtained from map data or through image recognition, including the number of floors of each building (large and small buildings); the density of vehicles on the street, which can be obtained by identifying the total number of vehicles on the street through image recognition and then calculating the vehicle density; the number of points of interest on the street, which can be obtained by obtaining an image of the street through image recognition and then identifying the points of interest in the image; the number of parking lots connected to the street, which can be obtained from map data; and the ratio of the historical invalid number of points of interest to the current number of points of interest (point of interest update rate of First Street or Second Street), which can be obtained from map data, including the historical invalid number and the current number of points of interest, and then the ratio of the two.

[0114] Similarly, when the target area type is the first street or the second street, the target area features can also have corresponding sample area features. For details, refer to the sample area features corresponding to the target area features of the first building. They will not be listed one by one here. In the end, we can also obtain a highly accurate point of interest density.

[0115] Step 304: Obtain the target interest point density corresponding to the target region features; wherein the target interest point density is calculated by a pre-trained regression model.

[0116] Step 305: Obtain the number of points of interest and the area size of the map data for the target area type.

[0117] Step 306: Calculate the number of interest points per unit size of the target region type based on the number of interest points and the region size, thereby obtaining the map interest point density corresponding to the target region feature.

[0118] In embodiments of the present invention, when the target area type is a building, the area size can represent the area, and the number of points of interest per unit area (e.g., per 100 square meters) is used as the actual point of interest density; when the target area type is a street, the area size can represent the length, and the number of points of interest per unit length (e.g., per 100 meters) is used as the actual point of interest density.

[0119] Step 307: Calculate the ratio of the map interest point density to the target interest point density. If the ratio is greater than the ratio threshold, it is determined that there is invalid interest point information in the map data of the target area type.

[0120] The embodiments of the present invention can quickly identify areas with a large number of invalid points of interest (POIs). After identifying areas with a large number of invalid POIs, invalidity detection can be further performed on individual POIs. For example, by calculating the similarity between two POIs within a certain distance and then combining the POI features, it can be determined whether the POIs are invalid. Alternatively, it can be determined whether the POIs are invalid by calculating behaviors associated with the POIs (such as WiFi connection behavior, payment behavior, etc.).

[0121] Optionally, after invalidating individual interest point information, the interest point density corresponding to the target region type is determined, which is the manually verified interest point density. The target region features corresponding to the target region type can be used as the sample region features corresponding to the target region type, and the manually verified interest point density corresponding to the target region type can be used as the sample interest point density corresponding to the target region type. The sample region features and sample interest point density corresponding to the target region type are used to optimize the regression model, which helps to improve the accuracy of the regression model in calculating the interest point density, making the detection of invalid interest points of the target region type more accurate.

[0122] Figure 4 This is a schematic diagram of an invalid point of interest detection device according to an embodiment of the present invention. Figure 4 As shown, the invalid interest point detection device 400 includes a first acquisition module 401, a second acquisition module 402, and a detection module 403. The first acquisition module 401 acquires target region features from map data of the target region type, thereby acquiring the target interest point density corresponding to the target region features. The target interest point density is calculated by a pre-trained regression model. The second acquisition module 402 acquires the map interest point density corresponding to the target region features. The detection module 403 calculates the ratio of the map interest point density to the target interest point density. If the ratio is greater than a ratio threshold, invalid interest point information is determined to exist in the map data of the target region type.

[0123] Optionally, a training module is also included for:

[0124] Obtain the sample region features and sample interest point density corresponding to the target region type;

[0125] The regression model is trained using the sample region features and the sample interest point density to obtain a trained regression model; wherein, the regression model is used to establish the mapping relationship between region features and interest point density;

[0126] The regression model is used to calculate the density of target interest points corresponding to the features of the target region.

[0127] Optionally, the second acquisition module 402 is further configured to:

[0128] Obtain the number of points of interest and the size of the target area's map data;

[0129] The number of interest points per unit size of the target region type is calculated based on the number of interest points and the region size, thereby obtaining the map interest point density corresponding to the target region feature.

[0130] Optionally, the first acquisition module 401 is further configured to:

[0131] Obtain the data to be detected for the target area in the map data;

[0132] The target region type of the target region is determined based on the data to be detected;

[0133] Obtain the target area features of the map data for the target area type.

[0134] Optionally, the target area features include at least one of the following: the city's tier, the city's region, and the ratio of the number of historical invalid points of interest to the number of current points of interest.

[0135] Optionally, if the target area feature is a region of the city where it is located, the first acquisition module 401 is further configured to:

[0136] The coverage rate of building area within the first preset range of the target area type is determined based on the map data;

[0137] If the coverage rate is greater than the first coverage rate threshold, then the target area feature is determined to be the feature of the central area of ​​the city.

[0138] If the coverage rate is less than the first coverage rate threshold and greater than the second coverage rate threshold, then the target area is determined to be a combination of the central area and the suburbs of the city.

[0139] If the coverage rate is less than the second coverage rate threshold, then the target area feature is determined to be a feature of the suburban area of ​​the city.

[0140] Optionally, the target area type includes at least one of a first street, a second street, and a first building; wherein the number of points of interest belonging to the service type in the first street is greater than the number of points of interest belonging to the service type in the second street;

[0141] If the target area type is the first street or the second street, then the target area characteristics include at least one of the following: vehicle density on the street, number of parking lots connected to the street, and average number of floors of buildings on the street.

[0142] If the target area type is the first building, then the target area features include at least one of the following: the number of floors in the first building, the number of points of interest in the first building, and the number of points of interest within a second preset range of the first building.

[0143] It should be noted that the specific implementation of the invalid interest point detection device of the present invention has been described in detail in the invalid interest point detection method described above, so the content will not be repeated here.

[0144] Figure 5 An exemplary system architecture 500 is shown, to which the invalid point of interest detection method or invalid point of interest detection apparatus of the present invention can be applied.

[0145] like Figure 5 As shown, system architecture 500 may include terminal devices 501, 502, and 503, a network 504, and a server 505. Network 504 serves as the medium for providing communication links between terminal devices 501, 502, and 503 and server 505. Network 504 may include various connection types, such as wired or wireless communication links, or fiber optic cables, etc.

[0146] Users can use terminal devices 501, 502, and 503 to interact with server 505 via network 504 to receive or send messages, etc. Various communication client applications can be installed on terminal devices 501, 502, and 503, such as shopping applications, web browser applications, search applications, instant messaging tools, email clients, social media platform software, etc. (for example only).

[0147] Terminal devices 501, 502, and 503 can be various electronic devices with displays that support web browsing, including but not limited to smartphones, tablets, laptops, and desktop computers.

[0148] Server 505 can be a server that provides various services, such as a backend management server that supports shopping websites browsed by users using terminal devices 501, 502, and 503 (for example only). The backend management server can analyze and process data such as received item information query requests, and then feed the processing results back to the terminal devices.

[0149] It should be noted that the invalid point of interest detection method provided in this embodiment of the invention is generally executed by server 505, and correspondingly, the invalid point of interest detection device is generally set in server 505.

[0150] It should be understood that Figure 5 The number of terminal devices, networks, and servers shown is merely illustrative. Depending on implementation needs, any number of terminal devices, networks, and servers can be included.

[0151] The following is for reference. Figure 6 It shows a schematic diagram of the structure of a computer system 600 suitable for implementing a terminal device of the present invention. Figure 6 The terminal device shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of the present invention.

[0152] like Figure 6 As shown, the computer system 600 includes a central processing unit (CPU) 601, which can perform various appropriate actions and processes based on programs stored in read-only memory (ROM) 602 or programs loaded from storage section 608 into random access memory (RAM) 603. The RAM 603 also stores various programs and data required for the operation of the system 600. The CPU 601, ROM 602, and RAM 603 are interconnected via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0153] The following components are connected to I / O interface 605: an input section 606 including a keyboard, mouse, etc.; an output section 607 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and speakers, etc.; a storage section 608 including a hard disk, etc.; and a communication section 609 including a network interface card such as a LAN card, modem, etc. The communication section 609 performs communication processing via a network such as the Internet. A drive 610 is also connected to I / O interface 605 as needed. A removable medium 611, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on drive 610 as needed so that computer programs read from it can be installed into storage section 608 as needed.

[0154] In particular, according to the embodiments disclosed in this invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments disclosed in this invention include a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication section 609, and / or installed from removable medium 611. When the computer program is executed by central processing unit (CPU) 601, it performs the functions defined above in the system of this invention.

[0155] It should be noted that the computer-readable medium shown in this invention can be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium can be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this invention, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In this invention, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. Computer-readable signal media can also be any computer-readable medium other than computer-readable storage media, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wireless, wire, optical fiber, RF, etc., or any suitable combination thereof.

[0156] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer programs according to various embodiments of the present invention. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0157] The modules described in the embodiments of the present invention can be implemented in software or hardware. The described modules can also be housed in a processor; for example, a processor can be described as including a first acquisition module, a second acquisition module, and a detection module, wherein the names of these modules do not necessarily limit the module itself.

[0158] In another aspect, the present invention also provides a computer-readable medium, which may be included in the device described in the above embodiments; or it may exist independently and not assembled into the device. The computer-readable medium carries one or more programs, and when the one or more programs are executed by the device, the device implements the following method: acquiring target region features of map data of a target region type, thereby acquiring the target interest point density corresponding to the target region features; wherein the target interest point density is calculated by a pre-trained regression model; acquiring the map interest point density corresponding to the target region features; calculating the ratio of the map interest point density to the target interest point density; if the ratio is greater than a ratio threshold, determining that invalid interest point information exists in the map data of the target region type.

[0159] In another aspect, embodiments of the present invention also provide a computer program product, including a computer program that, when executed by a processor, implements the methods described in any of the above embodiments.

[0160] According to the technical solution of this invention, by first acquiring the target area features of the map data of the target area type, thereby acquiring the target interest point density corresponding to the target area features, then acquiring the map interest point density corresponding to the target area features, and finally calculating the ratio of the map interest point density to the target interest point density, if the ratio is greater than a ratio threshold, it is determined that there is invalid interest point information in the map data of the target area type, thus overcoming the technical problem of low detection efficiency of invalid interest points in the prior art. This invention, by determining whether the ratio of the map interest point density to the target interest point density is greater than a ratio threshold, achieves the discovery of areas containing a large amount of invalid interest point information, that is, it achieves batch location of invalid interest point information, thereby improving the detection efficiency of invalid interest points.

[0161] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can occur depending on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A method for detecting invalid interest points, characterized in that, The method comprises the following steps: obtaining a target area feature of map data of a target area type, thereby obtaining a target point of interest density corresponding to the target area feature; wherein the target point of interest density is calculated by a pre-trained regression model; obtaining a map point of interest density corresponding to the target area feature; calculating a ratio of the map point of interest density and the target point of interest density, and if the ratio is greater than a ratio threshold, determining that there is invalid point of interest information in the map data of the target area type.

2. The method of claim 1, wherein, Before obtaining the target point of interest density corresponding to the target area feature, the method further comprises: obtaining a sample area feature and a sample point of interest density corresponding to the target area type; training the regression model by using the sample area feature and the sample point of interest density, thereby obtaining a trained regression model; wherein the regression model is used to establish a mapping relationship between an area feature and a point of interest density; calculating the target point of interest density corresponding to the target area feature by using the regression model.

3. The method of claim 1, wherein, Obtaining the map point of interest density corresponding to the target area feature comprises: obtaining a point of interest quantity and an area size of the map data of the target area type; calculating the number of points of interest per unit size of the target area type according to the point of interest quantity and the area size, thereby obtaining the map point of interest density corresponding to the target area feature.

4. The method of claim 1, wherein, Obtaining the target area feature of the map data of the target area type comprises: obtaining to-be-detected data of a target area in the map data; determining a target area type of the target area according to the to-be-detected data; obtaining the target area feature of the map data of the target area type.

5. The method of claim 1, wherein, The target area feature comprises at least one of a level of a city where the target area is located, a region of the city where the target area is located, a ratio of a historical number of invalid points of interest to a current number of points of interest.

6. The method of claim 5, wherein, If the target area feature is a region of the city where the target area is located, obtaining the target area feature of the map data of the target area type comprises: determining a coverage rate of a building area within a first preset range of the target area type according to the map data; if the coverage rate is greater than a first coverage rate threshold, determining that the target area feature is a feature of a central region of the city where the target area is located; if the coverage rate is less than the first coverage rate threshold and greater than a second coverage rate threshold, determining that the target area feature is a feature of a combined region of the central region and a suburban region of the city where the target area is located; if the coverage rate is less than the second coverage rate threshold, determining that the target area feature is a feature of a suburban region of the city where the target area is located.

7. The method of claim 1, wherein, The target area type comprises at least one of a first street, a second street, and a first building; wherein the number of points of interest of a service type in the first street is greater than the number of points of interest of the service type in the second street; if the target area type is the first street or the second street, the target area feature comprises at least one of a density of vehicles on the street, a number of parking lots connected to the street, and an average number of floors of buildings on the street. If the target region type is the first building, the target region feature includes at least one of a number of floors of the first building, a number of points of interest in the first building, and a number of points of interest within a second preset range of the first building.

8. A device for detecting invalid interest points, characterized in that Comprising: a first obtaining module, configured to obtain a target region feature of map data of a target region type, and to obtain a target point of interest density corresponding to the target region feature, wherein the target point of interest density is calculated by a pre-trained regression model; a second obtaining module, configured to obtain a map point of interest density corresponding to the target region feature; a detecting module, configured to calculate a ratio of the map point of interest density and the target point of interest density, and to determine that there is invalid point of interest information in the map data of the target region type if the ratio is greater than a ratio threshold.

9. An electronic device, comprising: Comprising: one or more processors; a storage device for storing one or more programs, when the one or more programs are executed by the one or more processors, the one or more processors implement the method of any one of claims 1-7.

10. A computer readable medium having stored thereon a computer program, characterized in that, The program is executed by the processor to implement the method of any one of claims 1-7.

11. A computer program product comprising a computer program, characterized in that, The computer program is executed by the processor to implement the method of any one of claims 1-7.

Images