CN111475746A

CN111475746A - Method and device for mining point of interest, computer equipment and storage medium

Info

Publication number: CN111475746A
Application number: CN202010265690.9A
Authority: CN
Inventors: 赵琳琳
Original assignee: Tencent Technology Shenzhen Co Ltd
Current assignee: Tencent Technology Shenzhen Co Ltd
Priority date: 2020-04-07
Filing date: 2020-04-07
Publication date: 2020-07-31
Anticipated expiration: 2040-04-07
Also published as: CN111475746B

Abstract

The application relates to a method and a device for mining a point of interest position based on big data, computer equipment and a storage medium. The method comprises the following steps: acquiring position information reported when resource transfer processing is carried out at an interest point to obtain a position information set; determining a reference area in a map where each piece of position information in the position information set is located; carrying out longitude and latitude grid division processing on the reference area to obtain a candidate area; screening a target area corresponding to the interest point from the candidate areas; the reporting quantity of the position information corresponding to the target area is higher than that of the position information corresponding to the non-target area; locating the location coordinates of the point of interest in the target area. By adopting the method, the efficiency of acquiring the interest point position can be improved.

Description

Method and device for mining point of interest, computer equipment and storage medium

Technical Field

The present application relates to the field of big data processing technology and the field of geographic information processing technology, and in particular, to a method and an apparatus for mining a point of interest, a computer device, and a storage medium.

Background

With the development of science and technology, big data analysis and processing are widely applied to various industries. The geographic information processing based on big data plays more and more important role in life and work of people. For a geographic information system, the information of the points of interest represents the value of the whole system to some extent. Therefore, it is important to acquire longitude and latitude information of the point of interest.

In the traditional method, a mapping staff needs to adopt a precise mapping instrument to measure the longitude and latitude of a point of interest and then mark the point of interest. It is obvious that the conventional method requires the position of the point of interest to be measured manually, resulting in very inefficient acquisition of the position of the point of interest.

Disclosure of Invention

In view of the above, there is a need to provide a method, an apparatus, a computer device and a storage medium for point of interest location mining, which can improve efficiency.

A method of point of interest location mining, the method comprising:

acquiring position information reported when resource transfer processing is carried out at an interest point to obtain a position information set;

determining a reference area in which each piece of position information in the position information set is located in a map;

carrying out longitude and latitude grid division processing on the reference area to obtain a candidate area;

screening a target area corresponding to the interest point from the candidate areas; the reporting quantity of the position information corresponding to the target area is higher than that of the position information corresponding to the non-target area;

the location coordinates of the point of interest are located in the target area.

A point of interest location mining apparatus, the apparatus comprising:

the acquisition module is used for acquiring the position information reported when the resource transfer processing is carried out at the interest point to obtain a position information set;

the area dividing module is used for determining a reference area in which each piece of position information in the position information set is located in the map; carrying out longitude and latitude grid division processing on the reference area to obtain a candidate area;

the screening module is used for screening a target area corresponding to the interest point from the candidate area; the reporting quantity of the position information corresponding to the target area is higher than that of the position information corresponding to the non-target area;

and the position coordinate positioning module is used for positioning the position coordinates of the interest points in the target area.

In one embodiment, the reference area is a longitude and latitude grid which is divided in advance on a map according to the preset longitude and latitude side length; the region dividing module is also used for positioning each position information in the position information set in the map; determining longitude and latitude grids in which the position information is positioned in the map to obtain a reference area; the reference area is at least one.

In one embodiment, the screening module is further configured to determine the reporting amount of the location information corresponding to each candidate area; filtering out candidate areas with the position information reporting quantity larger than or equal to a reporting quantity threshold value; a reporting quantity threshold value is determined according to the total quantity of the reporting quantities of the position information;

and screening the target area corresponding to the interest point from the filtered candidate areas.

In one embodiment, the screening module is further configured to perform extreme value detection processing on the position information report amount of the filtered candidate region to obtain an extreme value region; and determining the target area of the interest point according to the extreme value area.

In one embodiment, the points of interest are multiple points of interest belonging to the same object; the number of extremum regions is multiple; the screening module is further used for iteratively selecting a target extreme value region from the plurality of extreme value regions and selecting a neighborhood of the target extreme value region from the map according to a preset neighborhood selection condition; determining an extremum region located in the neighborhood to obtain a reference extremum region; when the target extremum region and the reference extremum region meet the report quantity approaching condition, judging that the target extremum region is the target region of the corresponding interest point; the reported quantity approaching condition is a preset condition indicating that the reported quantity of the position information corresponding to the target extremum region is close to the reported quantity of the position information corresponding to the reference extremum region.

In one embodiment, the screening module is further configured to sort the target extremum region and the reference extremum region in a descending order according to the reported amount of the position information; determining a dividing order according to the sorting result; the ratio of the position information reporting amount corresponding to the dividing order to the position information reporting amount corresponding to the previous dividing order is less than or equal to a preset threshold value; when the level of the target extreme value area is before the dividing level, the target extreme value area is judged to be the target area; and when the bit number of the target extreme value area is behind the demarcation bit number, judging that the target extreme value area is a non-target area.

In one embodiment, the screening module is further configured to select a current bit number in sequence from a first bit in the sorting result; and when the ratio of the position information reporting quantity corresponding to the next bit of the current bit to the position information reporting quantity corresponding to the current bit is greater than a preset threshold, taking the next bit as the current bit to perform iterative processing until the ratio is less than or equal to the preset threshold, and judging that the next bit of the current bit is the boundary bit.

In one embodiment, the screening module is further configured to obtain a preset radius value; and selecting a circular area on the map according to the radius value by taking the target extreme value area as the circle center to serve as the neighborhood of the target extreme value area.

In one embodiment, the radius value is multiple; the screening module is further used for continuing to select a neighborhood according to the next radius value for iterative processing when the reference extremum region in the neighborhood selected according to the last radius value meets the reported quantity approaching condition with the target extremum region, and judging the target extremum region as the target region of the corresponding interest point when the reference extremum region in the neighborhood selected according to the last radius value meets the reported quantity approaching condition with the target extremum region; the upper radius value is less than the lower radius value.

In one embodiment, the position coordinate positioning module is further configured to divide the target area to obtain a plurality of sub-areas; determining the position information reporting amount corresponding to each sub-area; and carrying out center point positioning on the sub-region with the most reported position information, and acquiring the longitude and latitude of the positioned center point to obtain the position coordinates of the interest point.

In one embodiment, the obtaining module is further configured to obtain a set of resource transfer data; the resource transfer data is data generated when the resource transfer processing is carried out on the interest points; each piece of resource transfer data carries position information reported when resource transfer processing is carried out; and respectively extracting the carried position information from each piece of resource transfer data in the resource transfer data set to obtain a position information set.

In one embodiment, the point of interest is an offline store, and the resource transfer process includes a payment process and location information, which is location information reported when the mobile terminal performs the payment process in the offline store.

A computer device comprising a memory storing a computer program and a processor, the processor implementing the steps of the method for point of interest location mining according to the embodiments of the present application when executing the computer program.

A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method for point of interest location mining according to embodiments of the present application.

The method, the device, the computer equipment and the storage medium for mining the position of the interest point acquire the position information reported when the resource transfer processing is carried out at the interest point, and obtain a position information set; and determining a reference area in the map where each piece of position information in the position information set is located. The reference area can embody an area range where the real position of the interest point is located. Then, carrying out longitude and latitude grid division processing on the reference area to obtain a candidate area; screening a target area corresponding to the interest point from the candidate areas; the reporting amount of the position information corresponding to the target area is higher than that of the position information corresponding to the non-target area. Namely, the region range where the real position is located is subjected to region subdivision and filtering screening, so that the region range where the real position of the interest point possibly exists is further reduced, and the target region where the interest point is most possibly located is screened out. Thus, the position coordinates of the point of interest can be accurately located in the target region. By utilizing the position information reported when the resource transfer processing is carried out on the interest point, the position coordinates of the interest point can be accurately mined and positioned, and compared with the traditional method that a professional is required to carry out manual measurement, the efficiency of obtaining the position of the interest point is greatly improved.

Drawings

FIG. 1 is a diagram of an embodiment of a method for point of interest location mining;

FIG. 2 is a diagram of an application environment of a point of interest location mining method in another embodiment;

FIG. 3 is a flowchart illustrating a method for point of interest location mining according to an embodiment;

fig. 4 is a schematic diagram illustrating reporting of location information in an embodiment;

FIG. 5 is a schematic diagram of the partitioning of candidate regions in one embodiment;

FIG. 6 is a diagram illustrating filtering candidate regions in one embodiment;

FIG. 7 is a diagram illustrating an extremum region in an embodiment;

FIG. 8 is a schematic illustration of cleaning extremum regions in one embodiment;

FIGS. 9-10 are schematic diagrams of interfaces for cleaning the extremum regions in one embodiment;

11-13 are diagrams illustrating the determination of coordinates of a point of interest location in one embodiment;

FIG. 14 is a block diagram of an apparatus for point of interest location mining in one embodiment;

FIG. 15 is a diagram showing an internal structure of a computer device according to an embodiment.

Detailed Description

In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.

The point of interest location mining method provided by the application can be applied to the application environment shown in fig. 1. Wherein the terminal 102 communicates with the server 104 via a network. The server 104 may be an independent physical server, a server cluster or a distributed system composed of a plurality of physical servers, or a cloud server providing basic cloud computing services such as a cloud service, a cloud database, cloud computing, a cloud function, cloud storage, a network service, cloud communication, middleware service, a domain name service, a security service, a CDN, a big data and artificial intelligence platform, and the like. . The terminal 102 may be, but is not limited to, a smart phone, a tablet, a laptop, a desktop computer, a smart speaker, a portable wearable device (e.g., a smart watch or smart glasses), and the like. The terminal and the server may be directly or indirectly connected through wired or wireless communication, and the application is not limited herein.

The terminal 102 has a resource transfer processing function and a positioning function. The terminal 102 may perform resource transfer processing at the point of interest, locate the terminal 102, and report the location information of the located terminal 102 to the server 104. It can be understood that the location information is reported during the resource transfer process of the point of interest, that is, the terminal 102 is located and reported at the point of interest.

The terminal 102 may be a mobile terminal (e.g., a handset having resource transfer functionality and location functionality). That is, a plurality of users use their respective terminals 102 to perform resource transfer processing at a point of interest, each terminal 102 locates its own position during the resource transfer processing, and reports the position information obtained by the positioning to the server 104. Then, the server 104 may obtain the location information reported when performing the resource transfer processing at the point of interest, to obtain a location information set.

The server 104 may determine a reference area in the map where each piece of location information in the set of location information is located; carrying out longitude and latitude grid division processing on the reference area to obtain a candidate area; screening a target area corresponding to the interest point from the candidate areas; the reporting quantity of the position information corresponding to the target area is higher than that of the position information corresponding to the non-target area; the location coordinates of the point of interest are located in the target area.

It is understood that the terminal 102 may also be a terminal that performs resource transfer processing and is fixedly disposed at a point of interest, for example, a fixed cash register device in an offline store.

It should be noted that fig. 1 illustrates a scene by taking location mining for only one point of interest as an example. In practice, multiple points of interest may be location mined in batches. I.e. the application scenario diagram as shown in fig. 2. Each terminal 102 performing resource transfer processing at a plurality of interest points reports location information to the server 104 during resource transfer processing. The location information set obtained by the server 104 includes location information reported when resource transfer processing is performed at a plurality of points of interest, so that the point of interest location mining method in the embodiments of the present application is executed based on the location information set to obtain a location coordinate of each point of interest.

In one embodiment, multiple points of interest may belong to the same object. Such as an off-line store belonging to the same brand.

It should be noted that, in other embodiments, other computer devices may also uniformly acquire the reported location information set from the server 104 and execute the method described in the embodiments of the present application, and the method described in the embodiments of the present application is not limited to be executed by the server 104 that directly receives the location information set reported by the terminal. It is understood that other computer devices may be terminals or backend servers (where backend servers are backend servers other than server 104). In addition, the terminal 102 may also directly report the location information to another terminal device that is equipped with the method for executing the embodiments of the present application, and is not limited to uploading to the server 102.

In one embodiment, as shown in fig. 3, a method for point of interest location mining is provided, which is described by taking the method as an example applied to the server in fig. 1, and includes the following steps:

step 302, obtaining the position information reported when the resource transfer processing is performed at the interest point, and obtaining a position information set.

A Point of Interest (POI) is a term in a geographic information system, and generally refers to any geographic object that can be abstracted as a Point. In the geographic information system, some geographic entities, such as a house, a shop, a mailbox, a bus station, etc., can be a point of interest.

The resource transfer process is a process for transferring resource data. The reported position information is the position information of the terminal reported when the terminal carries out resource transfer processing at the interest point. It can be understood that the location information is reported once for each resource transfer process. That is, the amount of reporting of the location information is the number of times the resource transfer process is performed. The reported position information is position information with longitude and latitude data, namely the position information belongs to the longitude and latitude information.

In one embodiment, the resource transfer process may include at least one of a payment process and a resource gifting process, among others. It will be appreciated that the amount paid in the payment process is in respect of the resource data, and that transferring the amount from one account to another is in respect of transferring the resource data.

It can be understood that the user can use the terminal to perform the resource transfer operation at the point of interest, and the terminal performs the resource transfer processing and locates itself to obtain the location information. The terminal can report the positioned position information to the server in the process of resource transfer processing. It can be understood that, at a point of interest, multiple users may perform multiple resource transfer processes, so that location information may be reported to a server multiple times, thereby forming a location information set.

In one embodiment, when performing the resource transfer process, the terminal may detect (i.e., sniff) a wireless network list (e.g., a Wi-Fi list), locate the location information of the terminal according to the wireless network list, and report the location information to the server during the resource transfer process. That is, the reported location information is obtained by positioning according to the detected wireless network list when the point of interest performs resource transfer processing.

Fig. 4 is a diagram illustrating reporting of location information according to an embodiment. Referring to fig. 4, considering factors such as instability of wireless network signals and interference from the outside (e.g., building blocking, flowing personnel), the positioning location information is inaccurate (i.e., the reported location information is not accurate enough) according to the wireless network list. For example, 10 mobile devices pay at the same convenience store, the 10 different location information a-J may be reported. As can be seen from fig. 4, the reported position information is somewhat far away from the real position of the convenience store, and is somewhat closer to the real position of the convenience store, so that the reported position information is not accurate enough, and accurate position coordinates (i.e., more accurate longitude and latitude) of the interest point (i.e., the convenience store) need to be mined based on the reported position information by the method in the embodiments of the present application.

In one embodiment, the point of interest is an offline store, and the resource transfer process includes a payment process and location information, which is location information reported when the mobile terminal performs the payment process in the offline store. That is, by the method in each embodiment of the present application, the position coordinates of the offline store are mined by using the position information reported when the online offline store performs payment processing by the mobile terminal.

In an embodiment, the server may directly obtain the location information that has been reported in the resource transfer processing stage, to obtain the location information set. That is, the location information may be reported in an independent form.

In other embodiments, the location information may also be reported in the form of resource transfer data together with other data generated during the resource transfer process. The server may extract the reported location information from the resource transfer data.

In one embodiment, step 302 includes: acquiring a set of resource transfer data; the resource transfer data is data generated when the resource transfer processing is carried out on the interest points; each piece of resource transfer data carries position information reported when resource transfer processing is carried out; and respectively extracting the carried position information from each piece of resource transfer data in the resource transfer data set to obtain a position information set.

Specifically, when the terminal performs the resource transfer processing at the point of interest, the terminal generates resource transfer data and reports the resource transfer data to the server. The reported resource transfer data includes the location information of the terminal when performing the resource transfer processing. The server can obtain a set of the reported resource transfer data, and then extract the position information carried in each resource transfer data to obtain a position information set.

For ease of understanding, this is now exemplified. For example, if 10 mobile terminals pay (i.e., pay, i.e., resource transfer operation) at the same convenience store (i.e., convenience store, i.e., point of interest), the location information reported 10 times can be obtained from the mobile payment data generated by 10 payments.

In one embodiment, the server may perform data cleaning on the location information set, remove location information with incomplete latitude information (e.g., missing field/incorrect format/scrambled code, etc.) in the mass data of the location information set, and perform step 304 and subsequent steps on the location information set after data cleaning.

Step 304, determining a reference area in the map where each piece of position information in the position information set is located.

The reference area is an area in the map for generalizing and representing the position of the interest point. It is understood that the reference region is used as a reference for locating the position of the interest point, i.e. the position coordinates of the located interest point can be further refined from the reference region. The reference area is at least one.

Specifically, the computer device may locate each piece of position information in the position information set on the map, and determine the reference area in which the position information set is located according to the location result.

In one embodiment, the reference area may be an area obtained by dividing the reference area in advance according to a preset rule, and the reference area into which each piece of location information falls is obtained by positioning each piece of location information on a map. In this case, the size of each reference area does not vary according to the position information sets, and the specific reference areas corresponding to different position information sets and the number of reference areas may vary, but do not affect the size of a single reference area, because the single reference area is divided according to a preset rule.

In one embodiment, the reference area may be generated by locating each piece of location information in the set of location information in real time. In this case, the size of the reference area is correlated with the position distribution located on the map based on each piece of position information. The sizes of the reference areas corresponding to different position information sets are different, for example, the more distributed the position information in the position information sets, the larger the corresponding reference area may be, and conversely, the more concentrated the position information in the position information sets, the smaller the corresponding reference area may be.

And step 306, performing longitude and latitude grid division processing on the reference area to obtain a candidate area.

The candidate area is a longitude and latitude grid obtained by dividing the longitude and latitude of the reference area.

Specifically, the server may further divide the longitude and latitude of the inside of each reference area, and use the divided longitude and latitude grid as a candidate area.

In one embodiment, the server may obtain a preset side length of the candidate longitude and latitude, and divide the interior of the reference area according to the side length of the candidate longitude and latitude to obtain a candidate area satisfying the side length of the candidate longitude and latitude. The candidate longitude and latitude side length may include a candidate longitude side length and a candidate latitude side length. The length of the candidate longitude side and the length of the candidate latitude side may be equal or different, and is not limited.

Fig. 5 is a schematic diagram illustrating the principle of dividing candidate regions in one embodiment. Referring to fig. 5, a longitude and latitude grid with a side length r of 100m is a reference area, and the reference area is subjected to longitude and latitude division according to a candidate longitude and latitude side length e of 20m to obtain a candidate area with a longitude and latitude side length of 20 m. That is, the reference area is divided into 25 candidate areas in fig. 5, each of which is a longitude and latitude grid having a side length of 20 m. And reporting the position information when the convenience store carries out mobile payment processing, wherein the number in each candidate area refers to the position information reporting amount corresponding to the candidate area, namely the number of times of reporting the position information in the candidate area. It can be understood that, in order to illustrate the relationship between the reported location information and the real location of the convenience store, the real location of the convenience store is marked by the location mark icon, and according to the method in the embodiments of the present application, the location coordinate (i.e., longitude and latitude coordinate) of the real location is obtained. As can be seen from fig. 5, the location information reported when the convenience store performs the mobile payment process is scattered around the real location of the convenience store, and therefore, the reported location information is non-precise longitude and latitude information scattered around the real location.

308, screening a target area corresponding to the interest point from the candidate areas; the reporting amount of the position information corresponding to the target area is higher than that of the position information corresponding to the non-target area.

The reporting amount of the position information is the number of times of reporting the position information. It can be understood that the reporting amount of the location information corresponding to the area is the number of times that the location information is reported in the area.

The target area refers to an area for locating the position of the interest point in the candidate areas. That is, the target region is the region where the true location of the point of interest is most likely located. The non-target region refers to a candidate region other than the target region.

In an embodiment, the server may determine the location information reporting amount corresponding to each candidate region, filter out candidate regions with a large location information reporting amount, and then select a target region corresponding to the interest point from the filtered candidate regions.

In another embodiment, the server may also determine the location information reporting amounts corresponding to the candidate areas, directly compare the location information reporting amounts, and select the candidate area with the highest location information reporting amount from the comparison result to obtain the target area.

As shown in fig. 5, the number in each candidate region in fig. 5 refers to the reported amount of the location information corresponding to the candidate region, i.e. the number of times the location information is reported in the candidate region. For example, the number 31 indicates the number of times of reporting the location information in the candidate area corresponding to the number 31, that is, the location information reported 31 times falls into the candidate area.

In step 310, the location coordinates of the point of interest are located in the target area.

It can be understood that the target area is an area which is most likely to have the real position of the interest point and is obtained by dividing and screening step by step from a large reference area in which the position information set is located in the map, which is equivalent to accurately mining the area in which the interest point may exist, and locating the position coordinates of the interest point.

In one embodiment, the server may further subdivide the target region, screen a sub-region with the largest amount of reported position information from the subdivided sub-regions, and then locate the position coordinates of the interest point from the screened sub-region.

In other embodiments, the server may directly locate the center point of the target area, and use the longitude and latitude coordinates of the center point as the position coordinates of the point of interest.

It should be noted that, when there are a plurality of interest points, the target area corresponding to each interest point can be obtained according to the steps 302 to 308, and then the position coordinates of the interest point corresponding to each target area are located in each target area. It will be appreciated that different points of interest may correspond to the same target area if their true locations are very close (e.g., the distance is less than a preset distance threshold).

The method for mining the position of the interest point obtains the position information reported when the resource transfer processing is carried out at the interest point, and obtains a position information set; and determining a reference area in the map where each piece of position information in the position information set is located. The reference area can embody an area range where the real position of the interest point is located. Then, carrying out longitude and latitude grid division processing on the reference area to obtain a candidate area; screening a target area corresponding to the interest point from the candidate areas; the reporting amount of the position information corresponding to the target area is higher than that of the position information corresponding to the non-target area. Namely, the region range where the real position is located is subjected to region subdivision and filtering screening, so that the region range where the real position of the interest point possibly exists is further reduced, and the target region where the interest point is most possibly located is screened out. Thus, the position coordinates of the point of interest can be accurately located in the target region. By utilizing the position information reported when the resource transfer processing is carried out at the interest point, the position coordinate of the interest point can be accurately mined and positioned. Compared with the traditional method which needs a professional to perform manual measurement, the method greatly improves the efficiency of obtaining the position of the interest point on the premise of ensuring the accuracy of the position coordinates of the excavated interest point. In addition, the labor cost is saved, and the cost is low.

In one embodiment, the reference area is a longitude and latitude grid pre-divided on a map according to a preset longitude and latitude side length. Step 304 includes: locating each piece of position information in the position information set in the map; determining longitude and latitude grids in which the position information is positioned in the map to obtain a reference area; the reference area is at least one.

It is understood that the predetermined longitude and latitude side lengths may include a longitude side length and a latitude side length. The longitude side length and the latitude side length may be equal or different, and are not limited. For example, the longitude and latitude may be divided by taking r1 ═ 0.01 latitude side length (corresponding to about 1.1 kilometer of the ground), and r2 ═ 0.01 longitude side length to obtain a longitude and latitude grid, i.e., a reference area. It is also possible to divide the latitude and longitude grid by letting r1 and r2 take different values.

Specifically, the server may divide the map into a plurality of latitude and longitude grids in advance according to the preset latitude and longitude side length, and each latitude and longitude grid may be used as a reference area. Thus, any positioning point that performs positioning on the map falls within the pre-divided longitude and latitude grid (i.e., the reference area). Then, the server may locate each piece of location information in the location information set in the map, to obtain a location point corresponding to each piece of location information. The server may determine the longitude and latitude grid where each location point is located, that is, determine a reference area where each location information is located after being located in the map. It is to be understood that the reference area in which the set of location information is located in the map is at least one.

As shown in fig. 5, the reported location information falls into the pre-divided reference area. And the latitude and longitude grids with numbers are regions with position information reported in the reference regions.

In the embodiment, the latitude grid division is performed on the map in advance to obtain the reference area, the position coordinates of the interest points are further positioned and calculated according to the reference area in which the position information set falls, which is equivalent to calculating the longitude and latitude coordinates of the interest points by using the latitude and longitude grid, and the efficiency of determining the position of the interest points is improved.

In one embodiment, step 306 includes: determining the position information reporting amount corresponding to each candidate area; filtering out candidate areas with the position information reporting quantity larger than or equal to a reporting quantity threshold value; a reporting quantity threshold value is determined according to the total quantity of the reporting quantities of the position information; and screening the target area corresponding to the interest point from the filtered candidate areas.

The total quantity of the reporting quantities of each piece of location information is the sum of the reporting quantities of the location information respectively corresponding to each candidate area.

Specifically, the server may determine the reporting amounts of the location information corresponding to the candidate areas, and sum the reporting amounts of the location information to obtain the total number of the reporting amounts of the location information. The server can determine a reporting quantity threshold according to the total quantity, and respectively compare the reporting quantities of the position information corresponding to the candidate areas with the reporting quantity threshold, and filter out the candidate areas with the reporting quantities of the position information being greater than or equal to the reporting quantity threshold. The server may filter a target region corresponding to the interest point from the filtered candidate regions.

In one embodiment, the server may obtain the adjustment factor, and determine the reporting amount threshold according to the total amount of the location information reporting amounts and the adjustment factor.

In one embodiment, the candidate regions may be filtered out according to the following formula:

wherein, c_EIndicating the reporting amount of location information corresponding to each candidate area as an adjustment factor ∑_E∈Rc_ERepresenting the total number of location information reports for each candidate area ∑_E∈Rc_EAnd expressing the reporting quantity threshold value. From the above formula, it can be seen that the reporting amount of the position information is greater than or equal to the threshold valueAnd filtering out the candidate areas for reservation, and discarding the candidate areas smaller than the report quantity threshold value.

The filtering of the candidate regions will now be explained in conjunction with fig. 5 and 6. Shown in fig. 5 is the candidate region before filtering. Fig. 6 shows the filtered candidate regions. Referring to fig. 6, candidate regions with a location information reporting amount greater than or equal to the reporting amount threshold are reserved, that is, the longitude and latitude grids other than 0 in fig. 6 (for example, the candidate regions corresponding to 235 and 821 are filtered and reserved candidate regions). The latitude and longitude grid labeled 0 in fig. 6 is the candidate area that is discarded (i.e., the candidate area with the location information reporting amount less than the reporting amount threshold). And then screening the target area corresponding to the interest point from each longitude and latitude grid (namely the filtered candidate area) which is not 0.

In the embodiment, the candidate area with a larger information reporting amount is filtered from the candidate areas according to the position information reporting amount, and the target area is determined based on the filtered candidate areas, so that the accuracy of position location is improved.

In one embodiment, screening the target region corresponding to the interest point from the filtered candidate regions comprises: carrying out extreme value detection processing on the filtered candidate areas, and detecting the candidate area with the largest position information reporting amount to obtain an extreme value area; and determining the target area of the interest point according to the extreme value area.

The extremum region refers to a candidate region with the largest information report amount in a local range in the filtered candidate regions.

Specifically, for any filtered candidate region, it is called an extremum region if the following conditions are satisfied: the reporting amount of the position information corresponding to the candidate area is larger than that of the position information corresponding to other filtered candidate areas at the periphery.

To facilitate understanding of the extremum regions, an explanation is now provided in connection with fig. 7. Referring to FIG. 7, a [ i, j ] is a candidate area in the ith row and jth column. If the filtered candidate area a [ i, j ] is larger than the position information reporting amount corresponding to the peripheral candidate areas (i.e. the candidate areas except a [ i, j ] in a [ i-1, j-1] -a [ i +1, j +1 ]), then the position information reporting amount corresponding to a [ i, j ] is the local maximum value, and the candidate area a [ i, j ] is the extremum area. It can be understood that the maximum amount of reported position information corresponding to the extremum region is equivalent to the maximum resource transfer processing amount (for example, the maximum transaction amount), and then the actual position of the point of interest is likely to be in the extremum region.

It should be noted that, there may be one or more extremum regions. When there is only one extremum region, the extremum region can be directly used as the target region of the interest point. When there are multiple extremal regions, the multiple extremal regions may be further cleaned and screened to determine a final target region. It is understood that when a point of interest is subjected to a location mining process, an extremum region may be obtained. When a lot of location mining is performed on multiple points of interest, multiple extremum regions may be obtained.

In the above embodiment, the extremum detection processing is performed on the filtered candidate regions, which is equivalent to further screening (i.e., secondary screening) the candidate regions in which the interest points are more likely to be located, so that the target region of the interest points is determined based on the screened regions, and the positioning accuracy is improved.

In one embodiment, the points of interest are multiple points of interest belonging to the same object; the extremum area is multiple. In this embodiment, determining the target region of the interest point according to the extremum region includes: iteratively selecting a target extremum region from the plurality of extremum regions, and selecting a neighborhood of the target extremum region from the map according to a preset neighborhood selection condition; determining an extremum region located in the neighborhood to obtain a reference extremum region; and when the target extremum region and the reference extremum region meet the report quantity approaching condition, judging that the target extremum region is the target region of the corresponding interest point.

The plurality of points of interest belonging to the same object means that the plurality of points of interest belong to the same object. For example, belonging to an off-line store under the same brand. The objects may be organizations, groups, brands, and the like, having a hierarchical architectural attribute.

It can be understood that the location information set includes location information reported when resource transfer processing is performed at each of a plurality of points of interest belonging to the same object. In the embodiment of the application, the position of the interest points belonging to the same object is uniformly mined by acquiring the position information reported when the resource transfer processing is respectively carried out on the interest points belonging to the same object. The method is equivalent to batch processing of a plurality of interest points belonging to the same object, so that a plurality of extremum regions are obtained. It can be understood that each interest point belonging to the same object has a corresponding extremum region, and the extremum region corresponding to each interest point is used for locating the final position coordinate of the interest point.

The preset neighborhood selection condition is a preset condition for selecting a neighborhood. And the target extremum region is an extremum region to be judged whether the target extremum region is the extremum region of the target region. The reference extremum region is an extremum region that is located in the neighborhood of the target extremum region, and is other than the target extremum region, among the obtained plurality of extremum regions. It is understood that the reference extremum region is used for reference when determining whether the target extremum region is the target region.

The reported quantity approaching condition is a preset condition indicating that the reported quantity of the position information corresponding to the target extremum region is close to the reported quantity of the position information corresponding to the reference extremum region.

Specifically, the server may iteratively select a target extremum region from a plurality of extremum regions corresponding to points of interest belonging to the same object, and select a neighborhood around the target extremum region from the map according to a preset neighborhood selection condition. The server may determine an extremum region located in the neighborhood, resulting in a reference extremum region. And when the target extremum region and the reference extremum region meet the report quantity approaching condition, judging that the target extremum region is the target region of the corresponding interest point. And when the report quantity approaching condition is not met between the target extreme value area and the reference extreme value area, judging that the target extreme value area is not the target area of the corresponding interest point, and rejecting the target extreme value area as a noise area.

It should be noted that, in the embodiment of the present application, each extremum region is sequentially used as a target extremum region, and the target extremum region is compared with a reference extremum region in its own neighborhood to determine whether the target extremum region is the target region. That is, it is necessary to perform neighborhood selection and difference comparison with a reference extremum region in its own neighborhood for each extremum region, and each processing result is only used to determine whether the extremum region itself is a target region, but not used to determine whether other extremum regions are target regions.

For example, if there are 3 extremum regions, a1 to a10, then a1 to a10 are sequentially targeted as the extremum regions. Assuming that when a1 is used as the target extremum region, the neighborhood of a1 includes 4 reference extremum regions a2 to a5, then, according to the difference in reported amount between a1 and a2 to a5, it is determined whether a1 is the target region (i.e., it is determined whether a1 needs to be reserved or removed). Assume a1 determines to cull. If a2 is the target extremum region, the neighborhood of a2 includes 5 reference extremum regions, i.e., a1 and a 3-a 6, and then, according to the reported difference between a2 and a1, a 3-a 6, it is determined whether a2 is the target region (i.e., it is determined whether a1 needs to be reserved or removed). Obviously, the judgment result of whether the a1 is reserved does not influence the judgment processing of the a 2.

In the above embodiment, each extremum region is compared with other reference extremum regions in the neighborhood to clean all extremum regions, so that the accuracy of data cleaning is improved, and the accuracy of subsequent position coordinates of interest points is improved.

In one embodiment, when the report approaching condition is satisfied between the target extremum region and the reference extremum region, determining that the target extremum region is the target region of the corresponding interest point includes: sorting the target extremum region and the reference extremum region in a descending order according to the reported amount of the position information; determining a dividing order according to the sorting result; the ratio of the position information reporting amount corresponding to the dividing order to the position information reporting amount corresponding to the previous dividing order is less than or equal to a preset threshold value; when the level of the target extreme value area is before the dividing level, the target extreme value area is judged to be the target area; and when the bit number of the target extreme value area is behind the demarcation bit number, judging that the target extreme value area is a non-target area.

It can be understood that the interest points corresponding to the target extremum region and the reference extremum region (i.e. the interest points corresponding to the respective extremum regions) belong to the same object. For example, when the point of interest is an offline store, the target extremum region and the reference extremum region may correspond to offline stores under the same brand. That is, the positions of multiple interest points belonging to the same object may be first generalized and represented by each extremum region, and then the extremum regions are cleaned by the method in the embodiment of the present application, and the extremum regions capable of more accurately representing the interest points are selected from the extremum regions. It is understood that the extremum region is not a specific location of the point of interest, but rather a generalized representation of the region in which the point of interest may exist.

The dividing level is a boundary of the reported amount of the position information and is equivalent to a watershed. The difference between the position information reporting amount corresponding to the bit before the division bit and the position information reporting amounts corresponding to the bit after the division bit satisfies the condition of excessive difference. That is, the reporting amount of the position information corresponding to the bit order before the division bit order is relatively large, and the reporting amount of the position information corresponding to the bit order after the division bit order is greatly reduced.

In one embodiment, a ratio of the position information reporting amount corresponding to the dividing level to the position information reporting amount corresponding to the previous level of the dividing level is less than or equal to a predetermined threshold.

Specifically, the server may sort the target extremum region and the reference extremum region in an order from a large reporting amount to a small reporting amount of the corresponding position information (i.e., a descending order). The server can determine the dividing order according to the sorting result; the ratio of the position information reporting amount corresponding to the dividing order to the position information reporting amount corresponding to the previous dividing order is less than or equal to a preset threshold value. It can be understood that, in this case, the difference in the reported amount of the location information between the bit before the dividing bit and the bit after the dividing bit is relatively large, that is, the condition of excessive difference is satisfied.

The server can determine the corresponding rank of the target extremum region in the sequence according to the sequencing result, when the rank of the target extremum region is before the dividing rank, the report quantity of the position information of the target extremum region is more, and the target extremum region can be judged to be the target region when the report quantity approaching condition is met between the server and other reference extremum regions. When the rank of the target extremum region is behind the dividing rank, the reported quantity of the position information of the target extremum region is very small, and the reported quantity is not close to other reference extremum regions, the target extremum region can be judged to be a non-target region, and the target extremum region can be used as a noise region to be removed.

In the above embodiment, for each target extremum region, the target extremum region and other reference extremum regions corresponding to the interest points in the neighborhood and belonging to the same object are sorted according to the reported amount of the position information, so as to determine the dividing rank, and since the dividing rank can accurately and intuitively divide the extremum regions with large difference in the reported amounts between the extremum regions in the sorting result, by determining whether the target extremum region is arranged before or after the dividing rank, it can be quickly and accurately determined whether the target extremum region is the target region, i.e., whether the target extremum region needs to be reserved or removed, thereby realizing quick cleaning of the extremum regions. Therefore, the accuracy and efficiency of data cleaning are improved. And the accuracy and the efficiency of the position coordinates of the subsequent positioning interest points are improved.

In one embodiment, determining the demarcation bit order according to the sorting result comprises: sequentially selecting the current bit from the first bit in the sequencing result; and when the ratio of the position information reporting quantity corresponding to the next bit of the current bit to the position information reporting quantity corresponding to the current bit is greater than a preset threshold, taking the next bit as the current bit to perform iterative processing until the ratio is less than or equal to the preset threshold, and judging that the next bit of the current bit is a boundary bit.

It can be understood that the sorting result is obtained by sorting the target extremum region and the reference extremum region in the neighborhood thereof in a descending order according to the reported amount of the position information. Therefore, the first extremum region in the sorting result is the extremum region with the largest amount of reported information in the extremum regions participating in the sorting.

The position information reporting amount corresponding to the rank is the position information reporting amount corresponding to the extremum region located in the rank.

Specifically, the server may select the current bit number in order starting from the first bit in the sorting result. And determining the ratio of the position information reporting quantity corresponding to the next time of the current time to the position information reporting quantity corresponding to the current time aiming at each current time.

It can be understood that when the ratio of the position information reporting amount corresponding to the next bit of the current bit to the position information reporting amount corresponding to the current bit is less than or equal to the preset threshold, it can be determined that the next bit of the current bit is the dividing bit.

When the ratio is greater than the preset threshold, it indicates that the reported amount of the position information corresponding to the next bit of the current bit is not much different from the reported amount of the position information corresponding to the current bit, and the next bit can be used as the new current bit to perform iterative processing, i.e., it is continuously determined whether the reported amount of the position information corresponding to the next bit of the new current bit is greater than the preset threshold or not to perform iterative processing until the ratio of the reported amount of the position information corresponding to the next bit of the current bit to the reported amount of the position information corresponding to the current bit is less than or equal to the preset threshold, and it is determined that the next bit of the current bit is the boundary bit.

For example, starting from the first position, when the ratio between the position information reporting amounts corresponding to the 2 nd order and the first position is greater than a preset threshold (for example, 0.1), the ratio between the position information reporting amounts corresponding to the 3 rd order and the 2 nd order is continuously compared with the preset threshold 0.1 until the iterative processing obtains that the ratio between the position information reporting amounts corresponding to the 5 th order and the 4 th order is less than the preset threshold 0.1, and then the 5 th order is the dividing order. Then, assuming that the rank of the target extremum region is before the 5 th rank, the target extremum region is retained (i.e. the target region), and assuming that the rank of the target extremum region is after the 5 th rank, the target extremum region is rejected. It can be understood that the reporting amount of the position information of the extremum region with the position rank after the 5 th position is much smaller than that of the extremum region with the position rank before, so that the noise region is likely to be, and the rejection can be performed.

In one embodiment, the cleaning process may be performed on the extremum region to determine the target region therefrom according to the following formula:

c₁＞c₂＞c₂…＞c_n

wherein, c₁,…,c_nRepresenting the reporting quantity of the position information corresponding to the n extreme value areas respectively; c. C₁＞c₂＞c₃…＞c_nThe n extremum regions are sorted in a descending order according to the reported quantity of the position information; t is the dividing order; t represents a preset threshold;

and the minimum bit which represents the condition that the ratio of the corresponding position information reporting quantity to the position information reporting quantity of the previous bit is less than a preset threshold value T is used as the dividing bit. c. C_tReporting the position information corresponding to the extremum region of the boundary order, c_iReporting the position information corresponding to the ith extremum region, c_i＞c_tThe reporting quantity of the position information representing the ith extremum region is greater than the reporting quantity of the position information corresponding to the demarcation rank, and it can be understood that, in this case, the rank of the extremum region is arranged before the demarcation rank; c. C_i≤c_tThe reporting quantity of the position information of the ith extreme value area is less than or equal to the reporting quantity of the position information corresponding to the demarcation bit order, in this case, the bit of the extreme value areaThe secondary row is after the demarcation bit; and f (a) is 1, the target extremum region is reserved, namely the target region, and the rank of the target extremum region is reserved before the dividing rank of the target extremum region as the target region. And f, if 0, the target extreme value area is removed, namely the target extreme value area is a non-target area, and the target extreme value area is removed after the bit rank of the target extreme value area is the dividing bit rank.

For ease of understanding, reference is now made to fig. 8 for illustration. Referring to fig. 8, if the predetermined threshold is 0.1, and the ratio between the reported amounts of the position information corresponding to the 5 th order and the 4 th order is less than the predetermined threshold 0.1 (i.e. the number of reported amounts of the position information corresponding to the extremum area ranked at the 5 th order is much less than that corresponding to the extremum area ranked at the 4 th order, which means that the number of transactions occurring in the extremum area ranked at the 5 th order is very small), then the 5 th order can be found. Assuming that the target extremum region is ranked before the 5 th order, for example, the target extremum region is ranked at the 3 rd order, i.e., the extremum region 3, then the target extremum region is retained (i.e., the target region). Assuming that the target extremum region is ranked at 5 th and later, for example, the target extremum region is ranked at 6 th, i.e. extremum region 6, then the target extremum region is rejected.

In the above embodiment, the iterative computation is started from the first position in the sorting result, so that the smallest position that satisfies the condition that the ratio between the corresponding position information reporting amount and the position information reporting amount of the previous position is smaller than the preset threshold value is used as the dividing position, the dividing position can be accurately determined, and the data does not need to be fully computed, so that the computation resource is saved and the efficiency is improved.

In one embodiment, selecting a neighborhood of the target extremum region from the map according to a preset neighborhood selection condition includes: acquiring a preset radius value; and selecting a circular area on the map according to the radius value by taking the target extreme value area as the circle center to serve as the neighborhood of the target extreme value area.

Specifically, the server may obtain a preset radius value, and select a circular area on the map with the target extremum area as a center of a circle and the obtained radius value as a radius, where the circular area is a neighborhood of the target extremum area. The server may obtain the extremum region located in the neighborhood to obtain a reference extremum region of the target extremum region.

Fig. 9-10 are schematic diagrams of interfaces for cleaning the extremum regions in an embodiment. Referring to fig. 9, if an extremum region a is a target extremum region, a circle is selected as a neighborhood region according to a radius r and a circle center is a circle center, and then an extremum region b is located in the neighborhood region. The information amount in fig. 9 and 10 is the position information report amount. Since a is much different from the reported quantity of the position information corresponding to b, the extremum area a can be determined to be a noise area (namely, a non-target area), and then the extremum area a can be cleaned. Fig. 10 is a schematic diagram of the cleaning of the extremum region a.

In one embodiment, the preset radius value is plural. When the target extremum region and the reference extremum region satisfy the report volume approaching condition, determining that the target extremum region is the target region of the corresponding interest point includes: when the reference extremum region in the neighborhood selected according to the last radius value meets the report quantity approaching condition with the target extremum region, continuing to select the neighborhood according to the next radius value for iterative processing until the reference extremum region in the neighborhood selected according to the last radius value meets the report quantity approaching condition with the target extremum region, and judging the target extremum region as the target region of the corresponding interest point; the upper radius value is less than the lower radius value.

It can be understood that, when there are a plurality of preset radius values, the server may select one radius value according to the order of the radius values from small to large, to perform the steps of selecting a circular area on the map as a neighborhood of the target extremum area by using the target extremum area as a center of the circle according to the radius values, and determining the extremum area located in the neighborhood to obtain the reference extremum area. And when the reference extremum region in the neighborhood selected according to the previous radius value meets the report quantity approaching condition with the target extremum region, continuously selecting the next radius value according to the sequence of the radius values from small to large, iteratively executing the steps of selecting a circular region on the map as the neighborhood of the target extremum region by taking the target extremum region as the center of a circle according to the radius values, and determining the extremum region in the neighborhood to obtain the reference extremum region. And judging that the target extreme value region is the target region of the corresponding interest point until the reference extreme value region in the neighborhood selected according to the last radius value meets the report quantity approaching condition with the target extreme value region.

For example, the predetermined radius values r are 50 meters, 100 meters and 500 meters, respectively. If it is determined whether the extremum region a needs to be reserved, the neighborhood 1 of the extremum region a may be selected by taking the extremum region a as a center of a circle and taking 50 meters as a radius. When the reference extremum region in the neighborhood 1 and the extremum region a do not satisfy the report approaching condition (for example, the ranking order of the extremum region a is behind the boundary order of the ranking result of the whole neighborhood 1), the extremum region a is determined to be a noise region, and the noise region needs to be removed and cleaned, so that the neighborhood is not selected according to the next radius value. When the reference extremum region and the extremum region a in the neighborhood 1 satisfy the reported quantity approaching condition (for example, the ranking order of the extremum region a is before the boundary order of the ranking result of the whole neighborhood 1), then the neighborhood 2 of the extremum region a can be selected continuously with the extremum region a as the center of circle and 100 meters as the radius, whether the reported quantity approaching condition is satisfied between the reference extremum region and the extremum region a in the neighborhood 2 is continuously judged, if so, the neighborhood 3 of the extremum region a can be selected continuously with the extremum region a as the center of circle and 500 meters as the radius, and when the reported quantity approaching condition is satisfied between the reference extremum region and the extremum region a in the neighborhood 3, the extremum region a can be finally judged as the target region and needs to be reserved.

In the embodiment, the data cleaning accuracy can be improved by selecting a plurality of radius values in sequence from small to large for layered cleaning.

In one embodiment, locating the location coordinates of the point of interest in the target region comprises: dividing a target area to obtain a plurality of sub-areas; determining the position information reporting amount corresponding to each sub-area; and carrying out center point positioning on the sub-region with the most reported position information, and acquiring the longitude and latitude of the positioned center point to obtain the position coordinates of the interest point.

Specifically, the server may further divide the inside of the target area according to a preset side length to obtain a plurality of sub-areas. It can be understood that, since the target area is a longitude and latitude grid, each sub-area obtained by dividing the target area is also equivalent to the longitude and latitude grid and has longitude and latitude information.

The server may determine the amount of location information reported corresponding to each sub-area, that is, determine the amount of location information reported in each sub-area. The server can compare the position information reporting quantity corresponding to each sub-region, determine the sub-region with the most position information reporting quantity, and perform center point positioning on the determined sub-region to obtain the center point of the sub-region. It can be understood that, since the sub-region is a longitude and latitude grid, the longitude and latitude corresponding to the central point of the sub-region can be obtained according to the longitude and latitude information corresponding to the determined sub-region.

11-13 are diagrams illustrating the determination of coordinates of a point of interest location in one embodiment. Referring to fig. 11, 25 sub-regions are obtained for the sub-regions into which the inside of the target region is divided. The number in each sub-region shown in fig. 12 is the reporting amount of the position information corresponding to the sub-region. A sub-area with the largest reported amount of the position information, that is, a sub-area with a display number of 107 (the number of the position information reported in the sub-area is 107), needs to be screened out. As shown in fig. 13, the server may center-point locate the sub-area displaying the number 107 to locate a center point (the black dot in fig. 13 is the center point), and calculate longitude and latitude (i.e., longitude: x, latitude: y) of the center point as the location coordinates of the convenience store. As can be seen from fig. 13, the sub-area with the number of 107 is displayed, that is, the sub-area with the largest amount of reported position information, and the closer to the sub-area, the larger the amount of reported position information, the more resource transfer processing is generated in the sub-area and its vicinity, and the more likely the real position of the convenience store is in the sub-area. As shown in fig. 13, the center point obtained by the final positioning is very close to the actual position of the convenience store, and therefore, the position coordinates of the interest point mined by the embodiment of the present application are very accurate.

It can be understood that, since the target region is obtained through multiple rounds of region division and region screening, the interest point is located in the corresponding target region to a large extent, that is, the target region can reflect the position of the interest point more accurately. And subdividing the target area again, and determining the sub-area with the largest position information reporting amount, so that the determined sub-area with the largest position information reporting amount can more accurately reflect the position of the interest point. Therefore, the real position of the interest point can be accurately reflected by taking the longitude and latitude of the determined central point as the position coordinate of the interest point corresponding to the target area.

In addition, the position coordinates of the interest points excavated by the method in the embodiments of the present application may also be applied to a recommendation algorithm based on L BS (L associated based Services), for example, when pushing the interest points to the user terminal, the position coordinates of the interest points may be accurately recommended to the user terminal according to the position coordinates of the excavated interest points.

It can be understood that the interest point location mining method in the embodiments of the present application is equivalent to mining the location of the interest point by performing big data analysis on the location information set by using an artificial intelligence technology.

Artificial Intelligence (AI) is a theory, method, technique and application system that uses a digital computer or a machine controlled by a digital computer to simulate, extend and expand human Intelligence, perceive the environment, acquire knowledge and use the knowledge to obtain the best results. In other words, artificial intelligence is a comprehensive technique of computer science that attempts to understand the essence of intelligence and produce a new intelligent machine that can react in a manner similar to human intelligence. Artificial intelligence is the research of the design principle and the realization method of various intelligent machines, so that the machines have the functions of perception, reasoning and decision making.

The artificial intelligence technology is a comprehensive subject and relates to the field of extensive technology, namely the technology of a hardware level and the technology of a software level. The artificial intelligence infrastructure generally includes technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing technologies, operation/interaction systems, mechatronics, and the like. The artificial intelligence software technology mainly comprises a computer vision technology, a voice processing technology, a natural language processing technology, machine learning/deep learning and the like.

It can be understood that the point of interest location mining method in the embodiments of the present application is equivalent to using a big data processing technology in an artificial intelligence technology.

Big data (Big data) refers to a data set which cannot be captured, managed and processed by a conventional software tool within a certain time range, and is a massive, high-growth-rate and diversified information asset which can have stronger decision-making power, insight discovery power and flow optimization capability only by a new processing mode. With the advent of the cloud era, big data has attracted more and more attention, and the big data needs special technology to effectively process a large amount of data within a tolerance elapsed time. The method is suitable for the technology of big data, and comprises a large-scale parallel processing database, data mining, a distributed file system, a distributed database, a cloud computing platform, the Internet and an extensible storage system. For example, the point of interest location mining method in embodiments of the present application may implement big data analysis processing based on a cloud computing platform, so as to mine the location of the point of interest.

It should be understood that, although the steps in the flowcharts are shown in sequence as indicated by the arrows, the steps are not necessarily performed in sequence as indicated by the arrows. The steps are not performed in the exact order shown and described, and may be performed in other orders, unless explicitly stated otherwise. Moreover, at least a part of the steps in each flowchart may include a plurality of steps or a plurality of stages, which are not necessarily performed at the same time, but may be performed at different times, and the order of performing the steps or stages is not necessarily sequential, but may be performed alternately or alternately with other steps or at least a part of the steps or stages in other steps.

In one embodiment, as shown in fig. 14, there is provided a point of interest location mining apparatus, which may be a part of a computer device using a software module or a hardware module, or a combination of the two, and specifically includes: an acquisition module 1402, an area division module 1404, a filtering module 1406, and a location coordinate locating module 1408, wherein:

an obtaining module 1402, configured to obtain location information reported when performing resource transfer processing at the point of interest, to obtain a location information set.

A region dividing module 1404, configured to determine a reference region in a map where each piece of location information in the location information set is located; and carrying out longitude and latitude grid division processing on the reference area to obtain a candidate area.

A screening module 1406, configured to screen, from the candidate regions, a target region corresponding to the interest point; the reporting amount of the position information corresponding to the target area is higher than that of the position information corresponding to the non-target area.

A location coordinate locating module 1408 for locating the location coordinates of the point of interest in the target area.

In one embodiment, the reference area is a longitude and latitude grid pre-divided on a map according to a preset longitude and latitude side length. The area division module 1404 is further configured to locate each piece of location information in the set of location information in the map; determining longitude and latitude grids in which the position information is positioned in the map to obtain a reference area; the reference area is at least one.

In one embodiment, the screening module 1406 is further configured to determine the reporting amount of the location information corresponding to each candidate area; filtering out candidate areas with the position information reporting quantity larger than or equal to a reporting quantity threshold value; a reporting quantity threshold value is determined according to the total quantity of the reporting quantities of the position information; and screening the target area corresponding to the interest point from the filtered candidate areas.

In one embodiment, the screening module 1406 is further configured to perform extremum detection on the reported amount of the position information of the filtered candidate region to obtain an extremum region; and determining the target area of the interest point according to the extreme value area.

In one embodiment, the points of interest are multiple points of interest belonging to the same object; the number of extremum regions is multiple; the screening module 1406 is further configured to iteratively select a target extremum region from the plurality of extremum regions, and select a neighborhood of the target extremum region from the map according to a preset neighborhood selection condition; determining an extremum region located in the neighborhood to obtain a reference extremum region; when the target extremum region and the reference extremum region meet the report quantity approaching condition, judging that the target extremum region is the target region of the corresponding interest point; the reported quantity approaching condition is a preset condition indicating that the reported quantity of the position information corresponding to the target extremum region is close to the reported quantity of the position information corresponding to the reference extremum region.

In one embodiment, the screening module 1406 is further configured to sort the target extremum regions and the reference extremum regions in a descending order according to the reported amount of the location information; determining a dividing order according to the sorting result; the ratio of the position information reporting amount corresponding to the dividing order to the position information reporting amount corresponding to the previous dividing order is less than or equal to a preset threshold value; when the level of the target extreme value area is before the dividing level, the target extreme value area is judged to be the target area; and when the bit number of the target extreme value area is behind the demarcation bit number, judging that the target extreme value area is a non-target area.

In one embodiment, the filtering module 1406 is further configured to sequentially select the current bit number from the first bit in the sorting result; and when the ratio of the position information reporting quantity corresponding to the next bit of the current bit to the position information reporting quantity corresponding to the current bit is greater than a preset threshold, taking the next bit as the current bit to perform iterative processing until the ratio is less than or equal to the preset threshold, and judging that the next bit of the current bit is the boundary bit.

In one embodiment, the filtering module 1406 is further configured to obtain a preset radius value; and selecting a circular area on the map according to the radius value by taking the target extreme value area as the circle center to serve as the neighborhood of the target extreme value area.

In one embodiment, the radius value is multiple. The screening module 1406 is further configured to, when the reference extremum region in the neighborhood selected according to the previous radius value and the target extremum region meet the reported quantity approaching condition, continue to select a neighborhood according to the next radius value for iterative processing, until the reference extremum region in the neighborhood selected according to the last radius value and the target extremum region meet the reported quantity approaching condition, determine that the target extremum region is the target region of the corresponding interest point; the upper radius value is less than the lower radius value.

In one embodiment, the location coordinate locating module 1408 is further configured to divide the target area into a plurality of sub-areas; determining the position information reporting amount corresponding to each sub-area; and carrying out center point positioning on the sub-region with the most reported position information, and acquiring the longitude and latitude of the positioned center point to obtain the position coordinates of the interest point.

In one embodiment, the obtaining module 1402 is further configured to obtain a set of resource transfer data; the resource transfer data is data generated when the resource transfer processing is carried out on the interest points; each piece of resource transfer data carries position information reported when resource transfer processing is carried out; and respectively extracting the carried position information from each piece of resource transfer data in the resource transfer data set to obtain a position information set.

For the specific definition of the point of interest location mining device, see the above definition of the point of interest location mining method, which is not described herein again. The modules in the above-mentioned interest point location mining apparatus may be implemented wholly or partially by software, hardware, and a combination thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.

In one embodiment, a computer device is provided, which may be a server or a terminal, and its internal structure diagram may be as shown in fig. 15. The computer device includes a processor, a memory, and a network interface connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a point of interest location mining method.

Those skilled in the art will appreciate that the architecture shown in fig. 15 is merely a block diagram of some of the structures associated with the disclosed aspects and is not intended to limit the computing devices to which the disclosed aspects may be applied, and that a particular computing device may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.

In one embodiment, a computer device is further provided, which includes a memory and a processor, the memory stores a computer program, and the processor implements the steps of the above method embodiments when executing the computer program.

In an embodiment, a computer-readable storage medium is provided, in which a computer program is stored which, when being executed by a processor, carries out the steps of the above-mentioned method embodiments.

It will be understood by those of ordinary skill in the art that all or part of the processes of the methods of the embodiments described above can be implemented by a computer program, which can be stored in a non-volatile computer-readable storage medium, and can include the processes of the embodiments of the methods described above when executed. Any reference to memory, storage, database or other medium used in the embodiments provided herein can include at least one of non-volatile and volatile memory. Non-volatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical storage, or the like. Volatile Memory can include Random Access Memory (RAM) or external cache Memory. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM), among others.

The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.

The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, and these are all within the scope of protection of the present application. Therefore, the protection scope of the present patent shall be subject to the appended claims.

Claims

1. A method for mining a position of interest, the method comprising:

determining a reference area in a map where each piece of position information in the position information set is located;

locating the location coordinates of the point of interest in the target area.

2. The method of claim 1, wherein the reference area is a longitude and latitude grid pre-divided on a map according to a preset longitude and latitude side length;

the determining a reference area in the map where each of the position information in the position information set is located includes:

locating each piece of location information in the set of location information in a map;

determining longitude and latitude grids in which the position information is positioned in the map to obtain a reference area; the reference area is at least one.

3. The method of claim 1, wherein the screening, from the candidate regions, a target region corresponding to the interest point comprises:

determining the position information reporting amount corresponding to each candidate area;

filtering out candidate areas with the position information reporting quantity being larger than or equal to a reporting quantity threshold value; the reporting quantity threshold is determined according to the total quantity of the reporting quantities of the position information;

and screening a target region corresponding to the interest point from the filtered candidate regions.

4. The method of claim 3, wherein the filtering the target region corresponding to the interest point from the filtered candidate regions comprises:

carrying out extreme value detection processing on the filtered position information reporting amount of the candidate region to obtain an extreme value region;

and determining a target area of the interest point according to the extreme value area.

5. The method of claim 4, wherein the points of interest are a plurality of points of interest belonging to the same object; the number of the extreme value areas is multiple; the determining the target region of the interest point according to the extremum region comprises:

iteratively selecting a target extremum region from a plurality of extremum regions, and selecting a neighborhood of the target extremum region from the map according to a preset neighborhood selection condition;

determining an extremum region located in the neighborhood to obtain a reference extremum region;

when the target extremum region and the reference extremum region meet the report quantity approaching condition, judging that the target extremum region is the target region of the corresponding interest point;

6. The method of claim 5, wherein when the report approaching condition is satisfied between the target extremum region and the reference extremum region, determining that the target extremum region is the target region of the corresponding interest point comprises:

sorting the target extremum region and the reference extremum region in a descending order according to the reported amount of the position information;

determining a dividing order according to the sorting result; the ratio of the position information reporting amount corresponding to the dividing order to the position information reporting amount corresponding to the previous dividing order is less than or equal to a preset threshold value;

when the rank of the target extreme value area is before the dividing rank, judging the target extreme value area as a target area;

and when the level of the target extreme value area is behind the dividing level, judging that the target extreme value area is a non-target area.

7. The method of claim 6, wherein determining the dividing order according to the sorting result comprises:

sequentially selecting the current bit from the first bit in the sequencing result;

and when the ratio of the position information reporting amount corresponding to the next bit of the current bit to the position information reporting amount corresponding to the current bit is greater than a preset threshold, taking the next bit as the current bit to perform iterative processing until the ratio is less than or equal to the preset threshold, and judging that the next bit of the current bit is a boundary bit.

8. The method of claim 5, wherein said selecting a neighborhood of the target extremum region from the map according to a predetermined neighborhood selection condition comprises:

acquiring a preset radius value;

and selecting a circular area on the map as a neighborhood of the target extreme value area according to the radius value by taking the target extreme value area as a circle center.

9. The method of claim 8, wherein the radius value is plural;

when the target extremum region and the reference extremum region satisfy the report volume approaching condition, determining that the target extremum region is the target region of the corresponding interest point includes:

when the reference extremum region in the neighborhood selected according to the last radius value meets the report quantity approaching condition with the target extremum region, continuing to select the neighborhood according to the next radius value for iterative processing until the reference extremum region in the neighborhood selected according to the last radius value meets the report quantity approaching condition with the target extremum region, and judging the target extremum region as the target region of the corresponding interest point; the upper radius value is less than the lower radius value.

10. The method of claim 1, wherein the locating the location coordinates of the point of interest in the target region comprises:

dividing the target area to obtain a plurality of sub-areas;

determining the position information reporting amount corresponding to each sub-area;

and carrying out center point positioning on the sub-region with the most reported position information, and acquiring the longitude and latitude of the positioned center point to obtain the position coordinate of the interest point.

11. The method of claim 1, wherein the obtaining the location information reported when the resource transfer processing is performed at the point of interest to obtain the location information set comprises:

acquiring a set of resource transfer data; the resource transfer data is data generated when the resource transfer processing is carried out on the interest points; each piece of resource transfer data carries position information reported when resource transfer processing is carried out;

and respectively extracting the carried position information from each piece of resource transfer data in the resource transfer data set to obtain a position information set.

12. The method according to any one of claims 1 to 11, wherein the point of interest is an offline store, the resource transfer process comprises a payment process, and the location information is location information reported by a mobile terminal when the mobile terminal performs the payment process in the offline store.

13. A point of interest location mining apparatus, the apparatus comprising:

the area dividing module is used for determining a reference area in which each piece of position information in the position information set is located in a map; carrying out longitude and latitude grid division processing on the reference area to obtain a candidate area;

the screening module is used for screening a target area corresponding to the interest point from the candidate areas; the reporting quantity of the position information corresponding to the target area is higher than that of the position information corresponding to the non-target area;

14. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor, when executing the computer program, implements the steps of the method of any of claims 1 to 12.

15. A computer-readable storage medium, in which a computer program is stored which, when being executed by a processor, carries out the steps of the method according to any one of claims 1 to 12.