WO2019128355A1

WO2019128355A1 - Method and device for determining accurate geographic location

Info

Publication number: WO2019128355A1
Application number: PCT/CN2018/108635
Authority: WO
Inventors: 肖明科
Original assignee: 北京京东尚科信息技术有限公司; 北京京东世纪贸易有限公司
Priority date: 2017-12-29
Filing date: 2018-09-29
Publication date: 2019-07-04
Also published as: CN109995884B; CN109995884A

Abstract

The present invention relates to the technical field of the Internet, and disclosed thereby are a method and device for determining an accurate geographic location. A preferred embodiment of the method comprises: acquiring an IP and a plurality of geographic locations associated with the IP; clustering the plurality of geographic locations by using a clustering algorithm, so as to obtain a geographic location clustering result of the IP; determining, according to the geographic location clustering result, an optimal geographic location corresponding to the IP by using an optimization algorithm; and determining a precise geographic location of the IP according to the optimal geographic location and a preset artificial neural network model. According to said preferred embodiment, the precise geographical location of the IP may be determined, thereby improving positioning accuracy without needing to install a large number of monitoring points, so that cost is reduced while the positioning accuracy is increased.

Description

Method and apparatus for determining precise geographic location

Technical field

The present application is based on a Chinese application with the application number of 201711481337.9 and the filing date is December 29, 2017, and the priority of which is hereby incorporated by reference.

The present invention relates to the field of Internet technologies, and in particular, to a method and apparatus for determining an accurate geographic location.

Background technique

IP positioning technology, in short, is a technology that determines the geographic location of a device by its IP address. IP positioning has an extremely wide range of applications, including targeted advertising, social networking, network security, performance optimization, and more. In the context of the mobile Internet, terminal devices including GPS information modules, such as mobile phones, can easily obtain the user's street-level geographic location through data reporting. However, if it is a terminal such as a desktop computer or a notebook that does not contain GPS hardware devices, it is impossible to obtain the user's geographic location through technologies such as GPS. In this case, high-precision IP positioning technology is required. The traditional IP positioning can only be located at the municipal level, and the accuracy of the district-level data is also debatable.

The traditional IP positioning algorithm estimates the position based on the linear relationship between the delay and the geographical distance, and reduces the error through the topology.

Specifically, it is based on BGP (Border Gateway Protocol)/ASN (Autonomous System Number) data analysis, and at the same time, the network monitoring point is built in the world, according to the IP to be located between the monitoring point and the monitoring point. The network return delay value divides the network topology to further confirm the geographical location of the IP to be located. This way, the positioning is more reliable, but the accuracy is still not high (zone level accuracy).

In the process of implementing the present invention, the inventors have found that at least the following problems exist in the prior art:

This technology needs to lay enough monitoring points to confirm the physical address of the IP, which is costly and requires more complicated steps, and because the network link delay is used to push back the geographical position, this way is more Trustworthy, but the accuracy is still not high.

Summary of the invention

In view of this, the embodiments of the present invention provide a method and apparatus for determining a precise geographic location, which improves positioning accuracy, and the present invention does not require a large number of monitoring points to be laid, thereby reducing the cost while improving positioning accuracy.

To achieve the above object, according to an aspect of an embodiment of the present invention, a method for determining an accurate geographic location includes: obtaining an IP and a plurality of geographic locations associated with the IP; using a clustering algorithm, Geographical clustering is performed to obtain a geographical location clustering result of the IP; and based on the geographical location clustering result, an optimal algorithm is used to determine an optimal geographic location corresponding to the IP; according to the optimal geographic location and pre- An artificial neural network model is set to determine the precise geographic location of the IP.

Optionally, the clustering algorithm is a k-means algorithm, and the optimization algorithm is a weighted least squares method.

Optionally, the step of clustering the plurality of geographic locations to obtain the geographical location clustering result of the IP by using a clustering algorithm comprises: selecting two geographical locations from multiple geographic locations associated with the IP a first initial centroid and a second initial centroid; calculating a first spherical distance between each of the plurality of geographic locations and the first initial centroid and a second spherical distance from the second initial centroid And clustering the plurality of geographical locations associated with the IP to obtain a high density cluster, and using the high density cluster as the geographic location cluster of the IP according to the first spherical distance and the second spherical distance result.

Optionally, a first spherical distance between each geographic location and the first initial centroid and a second spherical distance from the second initial centroid are calculated according to equation (1) below:

S=R·ar cos(cosβ1·cosβ2·cos(α1-α2)+sinβ1·sinβ2) (1)

Where R is the radius of the long axis of the earth, S is the spherical distance between the geographic location A and the geographic location B, β1 is the latitude of the geographic location A, α1 is the longitude of the geographic location A, β2 is the latitude of the geographic location B, and α2 is Longitude of location B.

Optionally, determining, according to the geographical location clustering result, an optimal algorithm for determining an optimal geographic location corresponding to each IP includes: for each geographic location in the high density cluster, according to each geographic location and a high density cluster centroid The spherical distance determines the weight of each of the geographic locations; according to the weights, the optimal geographic location corresponding to each IP is determined by a weighted least squares method.

Optionally, the weight of each of the geographic locations is determined according to the following formula (2):

Where λ _i represents the weight of the i-th geographic location, d _i represents the spherical distance between the i-th geographic location and the high-density cluster centroid, and n is an integer greater than or equal to 1;

Determining the optimal geographic location corresponding to the IP according to the following formula (3):

Where (x _i , y _i ) represents the ith geographic location,

Indicates the optimal geographical location.

Optionally, determining the precise geographic location of the IP according to the optimal geographic location and the preset artificial neural network model includes: inputting the optimal geographic location into the preset artificial neural network model, and obtaining an output. As a result; if the output result is a preset target result, the optimal geographic location is the precise geographic location of the IP.

Optionally, the input layer of the preset artificial neural network model has 3 neuron nodes, the hidden layer has 5 neuron nodes, and the output layer has 1 neuron node.

To achieve the above object, according to an aspect of an embodiment of the present invention, an apparatus for determining a precise geographic location is provided, including: an obtaining module, configured to acquire an IP and multiple geographic locations associated with the IP; a clustering module, The clustering algorithm is used to cluster the plurality of geographic locations to obtain a geographical location clustering result of the IP; an optimal geographic location determining module is configured to use an optimization algorithm based on the geographical location clustering result Determining an optimal geographic location corresponding to the IP; an accurate geographic location determining module, configured to determine an accurate geographic location of the IP according to the optimal geographic location and a preset artificial neural network model.

Optionally, the clustering module is further configured to: select two geographic locations from the plurality of geographic locations associated with the IP as the first initial centroid and the second initial centroid; calculate each of the multiple geographic locations a first spherical distance between the geographic location and the first initial centroid and a second spherical distance from the second initial centroid; the IP association based on the first spherical distance and the second spherical distance A plurality of geographical locations are clustered to obtain a high density cluster, and the high density cluster is used as a geographical location clustering result of the IP.

Optionally, the clustering module calculates a first spherical distance between each geographic location and the first initial centroid and a second spherical distance from the second initial centroid according to the following formula (1):

S=R·ar cos(cosβ1·cosβ2·cos(α1-α2)+sinβ1·sinβ2) (1)

Optionally, the optimal geographic location determining module is further configured to: determine, for each geographic location in the high density cluster, a weight of each geographic location according to a spherical distance of each geographic location and a high density cluster centroid According to the weight, the optimal geographic location corresponding to each IP is determined by a weighted least squares method.

Where (x _i , y _i ) represents the ith geographic location,

Indicates the optimal geographical location.

Optionally, the precise geographic location determining module is further configured to: input the optimal geographic location into the preset artificial neural network model, and obtain an output result; if the output result is a preset target result, The optimal geographic location is the precise geographic location of the IP.

To achieve the above object, according to an aspect of an embodiment of the present invention, an electronic device includes: one or more processors; and storage means for storing one or more programs when the one or more programs are Executed by the one or more processors, such that the one or more processors implement the method of determining an accurate geographic location as described in an embodiment of the present invention.

In order to achieve the above object, according to an aspect of an embodiment of the present invention, a computer readable medium storing a computer program, the program being executed by a processor to implement a determined precise geographic location as described in an embodiment of the present invention Methods.

One embodiment of the above invention has the following advantages or benefits: the clustering algorithm is used to cluster the plurality of geographic locations to obtain a geographical location clustering result for each IP; clustering results based on the geographic location Determining an optimal geographic location corresponding to the IP by using an optimization algorithm; determining a technical method of the precise geographic location of the IP according to the optimal geographic location and a preset artificial neural network model, thereby improving positioning accuracy, and There is no need to lay a large number of monitoring points, which reduces costs. Specifically, through k-means clustering, reduce redundant data, reduce GPS positioning errors caused by weather, signals, surrounding environment and other factors; then use weighted least squares method for different users (MAC) but the same IP geographical location Obtain the optimal geographic location; with the accumulation of data, establish an ANN neural network training model, train the optimal solution calculated by the same IP, and eliminate the dirty data caused by certain factors for a period of time (some mobile devices can Simulating GPS data, resulting in invalid GPS data), thereby improving the accuracy of positioning.

Further effects of the above-described non-conventional alternatives will be described below in connection with specific embodiments.

DRAWINGS

The drawings are intended to provide a better understanding of the invention and are not intended to limit the invention. among them:

1 is a schematic diagram of a main flow of a method of determining an accurate geographic location according to an embodiment of the present invention;

2 is a schematic diagram of a main flow of a method of determining an accurate geographic location according to another embodiment of the present invention;

3 is a schematic diagram of main modules of an apparatus for determining a precise geographic location, in accordance with an embodiment of the present invention;

4 is an exemplary system architecture diagram to which an embodiment of the present invention may be applied;

Figure 5 is a block diagram showing the structure of a computer system suitable for implementing a terminal device or server in accordance with an embodiment of the present invention.

Detailed ways

The exemplary embodiments of the present invention are described with reference to the accompanying drawings, and are in the Therefore, it will be apparent to those skilled in the art that various modifications and changes may be made to the embodiments described herein without departing from the scope and spirit of the invention. Also, descriptions of well-known functions and constructions are omitted in the following description for clarity and conciseness.

1 is a schematic diagram of a main flow chart of a method for determining an accurate geographic location of an IP-geographic data set in accordance with an embodiment of the present invention. As shown in Figure 1, the method includes:

Step S101: Obtain an IP and multiple geographical locations associated with the IP;

Step S102: Clustering the plurality of geographical locations by using a clustering algorithm to obtain a geographical location clustering result of the IP;

Step S103: Determine, according to the geographical location clustering result, an optimal geographic location corresponding to the IP by using an optimization algorithm;

Step S104: Determine an accurate geographic location of the IP according to the optimal geographic location and a preset artificial neural network model.

The IP in this embodiment and the plurality of geographic locations associated with the IP may be obtained through a public geographic information database. It can also be obtained by receiving the IP reported by the data collection source and multiple geographical locations associated with the IP, for example, receiving an IP address reported by a reporting device (for example, a smart phone, a tablet, etc.) having a GPS information module, and the IP address. The geographic location associated with the address.

With the development of the mobile Internet technology, any terminal device such as a mobile phone or a tablet computer can be used as a data collection source in the present embodiment. Therefore, the embodiment of the present invention does not need to lay a large number of monitoring points and reduces The cost.

In an optional embodiment, when receiving the IP reported by the reporting device and the geographical location associated with the IP, the device identifier (for example, a MAC address) of the reporting device and the time stamp when the data is reported may be acquired, thereby The device identification, time stamp, IP, and geographic location of the IP constitute a valid data, such as IP-MAC-GPS-TIMESTAMP, where GPS is the reported latitude and longitude information, and TIMESTAMP is the timestamp when the data is reported.

The above geographical location may be expressed as satellite positioning information such as latitude and longitude information, altitude information, or may be expressed as location information such as cities, streets, merchants, and office buildings. In this embodiment, the geographic location is preferably latitude and longitude information.

The above IP is essentially a 32-bit unsigned int data ranging from 0 to 2 ^32. For ease of use, the IP address in the form of a string is generally used, which is the usual 192.168.0.1. The form, in fact, converts every 8 binary bits into a corresponding decimal integer, abbreviated as a numeric IP. For example, 192.168.0.1 and 3232252721 are equivalent. 192.168.0.1 means 1*256 ⁰ +0*256 ¹ +168*256 ² +192*256 ³ =3232235521. In the embodiment of the present invention, the IP is a numerical IP for ease of use.

Since GPS is easily affected by factors such as the environment in which the reporting device is located, the weather, and the signal of the reporting device itself, errors in some geographical locations in the sample set may be large and cannot be fully accepted.

Therefore, for step S102, a plurality of geographical locations of the IP are clustered by using a clustering algorithm to exclude a geographical location with a large error, thereby obtaining a relatively accurate geographic location corresponding to the IP, thereby improving positioning accuracy. . As a specific example, the clustering algorithm may be a k-means clustering algorithm. Further, the device identifier may be used as a dimension and clustered by a timestamp, that is, the data reported by the same reporting device in a certain period of time is aggregated. class.

The above k-means algorithm is a typical distance-based clustering algorithm. The distance is used as the evaluation index of similarity, that is, the closer the distance between two objects is, the greater the similarity is. The core of the algorithm is to solve the problem by optimizing the distance from the data point to the centroid as a function of the optimization target, and using the function to take the extreme value to iterate continuously, so the compact and independent cluster is the final goal.

Further, as shown in FIG. 2, the step of clustering a plurality of geographical locations associated with the IP to obtain a geographical location clustering result of the IP by using a k-means clustering algorithm includes the following steps:

Step S201: Select two geographic locations from the plurality of geographic locations associated with the IP as the first initial centroid and the second initial centroid;

Step S202: calculating a first spherical distance between each of the plurality of geographical locations and the first initial centroid and a second spherical distance between the second initial centroid;

Step S203: Cluster the geographical locations associated with the IP according to the first spherical distance and the second spherical distance to obtain a high density cluster and a low density cluster, and use the high density cluster as the IP Geographic location clustering results.

For step S201, the latitude and longitude data collected for a period of time for the same reporting device (ie, the same IP) is hashed near the real geographical location of the IP, and such points are dense, but due to external factors Influence, a few points have a large deviation from the real position, and the density is sparse. Therefore, the embodiment of the present invention defines clusters as high-density regions separated by low-density regions. When the initial centroid is selected, two types are selected on the density-based clusters.

Specifically, two latitude and longitude may be randomly selected as the first initial centroid and the second initial centroid, or the average of all the latitude and longitude may be selected as the first initial centroid, and the latitude and longitude with the largest deviation from the average is taken as the second initial centroid.

For step S202, since the latitude and longitude is the coordinates of the ellipsoid, the Euclidean distance cannot be simply used as a compact index for measuring the cluster, and the embodiment of the present invention uses the spherical distance as a compact index for measuring the cluster. The spherical distance between two geographic locations can be calculated by the following formula:

S=R·ar cos(cosβ1·cosβ2·cos(α1-α2)+sinβ1·sinβ2) (1)

For the step S203, after the first spherical distance and the second spherical distance are calculated according to the formula (1), the geographical position close to the first initial centroid is a cluster, and the geographical position close to the second initial centroid is another cluster. Then, recalculate the centroid of each cluster and repeat the iteration until the final centroid is constant or the change is small. The high-density cluster is selected as the geographical clustering result of the IP, and the low-density cluster is discarded as the error cluster to avoid data pollution.

For step S103, after the clustering algorithm preliminarily excludes the geographical location with large error, in order to further improve the positioning accuracy, an optimization algorithm is needed to determine the optimal geographical position corresponding to each IP. Specifically, an optimization algorithm can be used to obtain an optimal solution for a high-density cluster of the same IP. As a specific example, the optimization algorithm may be a weighted least squares method.

The weighted least squares method described above is a mathematical optimization technique that finds the best function match of the data by minimizing the sum of the squares of the errors. The weighted least squares method has a wide range of applications in the field of engineering technology. The weighted least squares method can be used to easily obtain unknown parameters and minimize the sum of squared errors between these obtained data and actual data.

Specifically, the process of determining the optimal geographic location corresponding to the IP by using a weighted least squares method based on the geographic location clustering result may include the following steps:

1. For each geographic location in the high density cluster, determining the weight of each geographic location based on the spherical distance of each geographic location and the high density cluster centroid;

The formula is as follows:

λ _i represents the weight of the i-th latitude and longitude, d _i represents the distance between the i-th latitude and longitude and the centroid, and n is an integer greater than or equal to 1.

2. Based on the weights, the weighted least squares method is used to determine the optimal geographic location corresponding to each IP. In this process, it is necessary to establish a nonlinear curve fitting function for the latitude and longitude of the same IP to minimize the variance. The specific formula is as follows: (3):

Where (x _i , y _i ) represents the ith geographic location,

The optimal geographical location corresponding to the IP. In the actual calculation, (x _i , y _i ) is the plane coordinate after the latitude and longitude is converted to the geodetic coordinates by the Gauss projection by the ith geographic location.

In the embodiment of the present invention, a nonlinear regression model is established for the latitude and longitude data of the same IP:

among them

For the center coordinates, r is the radius. Find the optimal geographic location corresponding to the IP

Make it satisfy

The smallest.

For step S103, after the k-means algorithm and the weighted least squares method described above, it can be determined that the data reported by a sampling device has been correctly processed, but in the actual process, the reported IP and latitude and longitude data are present due to factors such as a simulator. There may be large deviations, and this part of the data can be considered as abnormal data. Therefore, in the present embodiment, an artificial neural network model can be utilized to filter the optimal geographic location calculated by the same IP, thereby eliminating abnormal data. Specifically, after determining the optimal geographic location of the IP, an artificial neural network model is introduced to perform a simple 'classification' on the optimal geographic location, that is, all the optimal geographic locations are divided into two categories, normal and abnormal. class.

Therefore, further, the method further comprises: determining an accurate geographic location of the IP according to the optimal geographic location and a preset artificial neural network model.

Specifically, the following steps may be included:

Inputting the optimal geographic location into the preset artificial neural network model to obtain an output result;

If the output result is a preset target result, the optimal geographic location is an exact geographic location of the IP.

Before screening the optimal geographic location using the preset artificial neural network model, the method further comprises: training the artificial neural network model, that is, adjusting the weight of each neural node through the training data, so that the expected output of the normal optimal geographic location is obtained. For 1, the expected output of the abnormally optimal geographic location is zero.

Specifically, a plurality of IP data associated with the correct geographical location are selected as normal data (for example, greater than 20,000 data), and artificial abnormal data is added to the same IP, and the artificial neural network model hidden layer weight training is performed by using the normal data and the artificial abnormal data. The final function is guaranteed to converge, and the hidden layer weight parameter is used as the initialization parameter.

As a specific example, the input layer of the preset artificial neural network model has three neuron nodes corresponding to IP (numerical IP), longitude and latitude; the hidden layer has five neuron nodes, and the number of nodes It is determined by the developer through the training data convergence time and method; the output layer has one neuron node, and the output result is used to determine whether the latitude and longitude is abnormal data, the output result is 1 indicating that the latitude and longitude is normal data, and the output result is 0 indicating the latitude and longitude. For abnormal data.

Therefore, the above-mentioned preset target result may be 1, and if the output result is 1, the optimal geographical position is the precise geographical position of the IP.

In an alternative embodiment, the obtained IP and the precise geographic location of the IP may be saved.

In this embodiment, the Artificial Neural Network (ANN) is: abstracting the human brain neural network from the perspective of information processing, establishing a simple model, and forming different networks according to different connection modes. In engineering and academia, it is often referred to directly as a neural network or a neural network. A neural network is an operational model consisting of a large number of nodes (or neurons) connected to each other. Each node represents a specific output function called an activation function. The connection between every two nodes represents a weighting value for passing the connection signal, called weight, which is equivalent to the memory of the artificial neural network. The output of the network varies depending on the connection method of the network, the weight value and the excitation function. The network itself is usually an approximation of an algorithm or function in nature, or it may be an expression of a logic strategy.

The method for determining the precise geographical location of the embodiment of the invention improves the positioning accuracy, and does not need to lay a large number of monitoring points, thereby reducing the cost. Specifically, through k-means clustering, reduce redundant data, reduce GPS positioning errors caused by weather, signals, surrounding environment and other factors; then use weighted least squares method for different users (MAC) but the same IP geographical location Obtain the optimal geographic location; with the accumulation of data, establish an ANN neural network training model, train the optimal solution calculated by the same IP, and eliminate the dirty data caused by certain factors for a period of time (some mobile devices can Simulating GPS data, resulting in invalid GPS data), thereby improving the accuracy of positioning.

The method of the embodiment of the present invention can also obtain the variance according to the formula (3), and provide a quantitative indicator for the accuracy of the IP positioning. The smaller the variance, the higher the accuracy.

3 is a schematic diagram of main modules of an IP positioning apparatus according to still another embodiment of the present invention. As shown in FIG. 3, the apparatus 300 includes: an obtaining module 301, configured to acquire an IP and multiple geographic locations associated with the IP; and a clustering module 302, configured to use the clustering algorithm to perform the multiple geographic locations Performing clustering to obtain a geographical location clustering result of the IP; an optimal geographic location determining module 303, configured to determine, according to the geographic location clustering result, an optimal geographic location corresponding to the IP by using an optimization algorithm; The location determining module 304 is configured to determine an accurate geographic location of the IP according to the optimal geographic location and a preset artificial neural network model.

Optionally, the clustering module 302 is further configured to: select two geographic locations from the plurality of geographic locations associated with the IP as the first initial centroid and the second initial centroid; calculate each of the multiple geographic locations a first spherical distance between the geographic location and the first initial centroid and a second spherical distance from the second initial centroid; the IP according to the first spherical distance and the second spherical distance The associated plurality of geographic locations are clustered to obtain a high density cluster, and the high density cluster is used as a geographical location clustering result of the IP.

Optionally, the clustering module 302 calculates a first spherical distance between each geographic location and the first initial centroid and a second spherical distance from the second initial centroid according to the following formula (1):

S=R·ar cos(cosβ1·cosβ2·cos(α1-α2)+sinβ1·sinβ2) (1)

Optionally, the optimal geographic location determining module 303 is further configured to: determine, for each geographic location in the high density cluster, the geographic distance of each geographic location and the high density cluster centroid, determine each geographic location Weights; based on the weights, the weighted least squares method is used to determine the optimal geographic location corresponding to each IP.

Where (x _i , y _i ) represents the ith geographic location,

Indicates the optimal geographical location.

Optionally, the precise geographic location determining module 304 is further configured to: input the optimal geographic location into the preset artificial neural network model, and obtain an output result; if the output result is a preset target result, The optimal geographic location is then the precise geographic location of the IP.

The device for determining the precise geographical position of the embodiment of the invention improves the positioning accuracy, and does not need to lay a large number of monitoring points, thereby reducing the cost. Specifically, through k-means clustering, reduce redundant data, reduce GPS positioning errors caused by weather, signals, surrounding environment and other factors; then use weighted least squares method for different users (MAC) but the same IP geographical location Obtain the optimal geographic location; with the accumulation of data, establish an ANN neural network training model, train the optimal solution calculated by the same IP, and eliminate the dirty data caused by certain factors for a period of time (some mobile devices can Simulating GPS data, resulting in invalid GPS data), thereby improving the accuracy of positioning.

4 illustrates an exemplary system architecture 400 of an IP-geographic data set construction method or IP-geographic data set construction apparatus to which embodiments of the present invention may be applied.

As shown in FIG. 4, system architecture 400 can include

terminal devices

401, 402, 403, network 404, and server 405. Network 404 is used to provide a medium for communication links between

terminal devices

401, 402, 403 and server 405. Network 404 can include a variety of connection types, such as wired, wireless communication links, fiber optic cables, and the like.

The user can interact with the server 405 via the network 404 using the

terminal devices

401, 402, 403 to receive or send messages and the like. The

terminal devices

401, 402, 403 may be various electronic devices having a display screen and supporting web browsing, including but not limited to smart phones, tablet computers, laptop portable computers, desktop computers, and the like.

The server 405 may be a server that provides various services, such as a background management server that provides support to a shopping site browsed by the user using the

terminal devices

401, 402, and 403. The background management server may analyze and process data such as the received product information query request, and feed back the processing result (for example, target push information and product information) to the terminal device.

It should be noted that the method for determining the precise geographic location provided by the embodiment of the present invention is generally performed by the server 405. Accordingly, the IP positioning device is generally disposed in the server 405.

It should be understood that the number of terminal devices, networks, and servers in FIG. 4 is merely illustrative. Depending on the implementation needs, there can be any number of terminal devices, networks, and servers.

Referring now to Figure 5, there is shown a block diagram of a computer system 500 suitable for use in implementing a terminal device in accordance with an embodiment of the present invention. The terminal device shown in FIG. 5 is merely an example, and should not impose any limitation on the function and scope of use of the embodiments of the present invention.

As shown in FIG. 5, computer system 500 includes a central processing unit (CPU) 501 that can be loaded into a program in random access memory (RAM) 503 according to a program stored in read only memory (ROM) 502 or from storage portion 508. And perform various appropriate actions and processes. In the RAM 503, various programs and data required for the operation of the system 500 are also stored. The CPU 501, the ROM 502, and the RAM 503 are connected to each other through a bus 504. An input/output (I/O) interface 505 is also coupled to bus 504.

The following components are connected to the I/O interface 505: an input portion 506 including a keyboard, a mouse, etc.; an output portion 507 including, for example, a cathode ray tube (CRT), a liquid crystal display (LCD), and the like, and a storage portion 508 including a hard disk or the like. And a communication portion 509 including a network interface card such as a LAN card, a modem, or the like. The communication section 509 performs communication processing via a network such as the Internet. Driver 510 is also coupled to I/O interface 505 as needed. A removable medium 511 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory or the like is mounted on the drive 510 as needed so that a computer program read therefrom is installed into the storage portion 508 as needed.

In particular, the processes described above with reference to the flowcharts may be implemented as a computer software program in accordance with embodiments of the present disclosure. For example, embodiments of the present disclosure include a computer program product comprising a computer program embodied on a computer readable medium, the computer program comprising program code for executing the method illustrated in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network via the communication portion 509, and/or installed from the removable medium 511. When the computer program is executed by the central processing unit (CPU) 501, the above-described functions defined in the system of the present invention are performed.

It should be noted that the computer readable medium shown in the present invention may be a computer readable signal medium or a computer readable storage medium or any combination of the two. The computer readable storage medium can be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples of computer readable storage media may include, but are not limited to, electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read only memory (ROM), erasable Programmable read only memory (EPROM or flash memory), optical fiber, portable compact disk read only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination of the foregoing. In the present invention, a computer readable storage medium may be any tangible medium that can contain or store a program, which can be used by or in connection with an instruction execution system, apparatus or device. In the present invention, a computer readable signal medium may include a data signal that is propagated in the baseband or as part of a carrier, in which computer readable program code is carried. Such propagated data signals can take a variety of forms including, but not limited to, electromagnetic signals, optical signals, or any suitable combination of the foregoing. The computer readable signal medium can also be any computer readable medium other than a computer readable storage medium, which can transmit, propagate, or transport a program for use by or in connection with the instruction execution system, apparatus, or device. . Program code embodied on a computer readable medium can be transmitted by any appropriate medium, including but not limited to wireless, wire, optical cable, RF, etc., or any suitable combination of the foregoing.

The flowchart and block diagrams in the Figures illustrate the architecture, functionality and operation of possible implementations of systems, methods and computer program products in accordance with various embodiments of the invention. In this regard, each block of the flowchart or block diagrams can represent a module, a program segment, or a portion of code that includes one or more Executable instructions. It should also be noted that in some alternative implementations, the functions noted in the blocks may also occur in a different order than that illustrated in the drawings. For example, two successively represented blocks may in fact be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending upon the functionality involved. It is also noted that each block of the block diagrams or flowcharts, and combinations of blocks in the block diagrams or flowcharts, can be implemented by a dedicated hardware-based system that performs the specified function or operation, or can be used A combination of dedicated hardware and computer instructions is implemented.

The modules involved in the embodiments of the present invention may be implemented by software or by hardware. The described modules may also be disposed in a processor, for example, as a processor including a transmitting module, an obtaining module, a determining module, and a first processing module. The name of these modules does not constitute a limitation on the unit itself in some cases. For example, the sending module may also be described as a module that sends a picture acquisition request to the connected server.

In another aspect, the present invention also provides a computer readable medium, which may be included in the apparatus described in the above embodiments, or may be separately present and not incorporated in the apparatus. The computer readable medium carries one or more programs, and when the one or more programs are executed by the device, the device includes: obtaining an IP and a plurality of geographic locations associated with the IP; using a clustering algorithm And clustering the plurality of geographic locations to obtain a geographical location clustering result of the IP; determining, according to the geographic location clustering result, an optimal geographic location corresponding to the IP by using an optimization algorithm; An excellent geographic location and a preset artificial neural network model to determine the precise geographic location of the IP.

The technical solution of the embodiment of the present invention uses a clustering algorithm to cluster the plurality of geographical locations to obtain a geographical location clustering result of each IP; and based on the geographical location clustering result, determine the The optimal geographic location corresponding to the IP; the technical means for determining the precise geographical location of the IP according to the optimal geographic location and the preset artificial neural network model, so the positioning accuracy is improved, and a large number of monitoring points are not required to be laid. Reduced costs. Specifically, through k-means clustering, reduce redundant data, reduce GPS positioning errors caused by weather, signals, surrounding environment and other factors; then use weighted least squares method for different users (MAC) but the same IP geographical location Obtain the optimal geographic location; with the accumulation of data, establish an ANN neural network training model, train the optimal solution calculated by the same IP, and eliminate the dirty data caused by certain factors for a period of time (some mobile devices can Simulating GPS data, resulting in invalid GPS data), thereby improving the accuracy of positioning.

The above specific embodiments do not constitute a limitation of the scope of the present invention. Those skilled in the art will appreciate that a wide variety of modifications, combinations, sub-combinations and substitutions can occur depending on design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and scope of the invention are intended to be included within the scope of the invention.

Claims

A method for determining an accurate geographic location, comprising:

Obtaining an IP and a plurality of geographic locations associated with the IP;

Clustering the plurality of geographic locations to obtain a geographical location clustering result of the IP by using a clustering algorithm;

Determining, according to the geographical location clustering result, an optimal geographic location corresponding to the IP by using an optimization algorithm;

Determining the precise geographic location of the IP based on the optimal geographic location and a preset artificial neural network model.
The method according to claim 1, wherein the clustering algorithm is a k-means algorithm, and the optimization algorithm is a weighted least squares method.
The method according to claim 2, wherein the step of clustering the plurality of geographical locations to obtain the geographical location clustering result of the IP by using a clustering algorithm comprises:

Selecting two geographic locations from the plurality of geographic locations associated with the IP as the first initial centroid and the second initial centroid;

Calculating a first spherical distance between each of the plurality of geographic locations and the first initial centroid and a second spherical distance from the second initial centroid;

And clustering a plurality of geographical locations associated with the IP to obtain a high density cluster, and using the high density cluster as a geographical location clustering result of the IP according to the first spherical distance and the second spherical distance .
The method according to claim 3, wherein the first spherical distance between each geographic location and the first initial centroid and the second spherical surface between the second initial centroid are calculated according to the following formula (1) distance:

S=R·ar cos(cosβ1·cosβ2·cos(α1-α2)+sinβ1·sinβ2) (1)

Where R is the radius of the long axis of the earth, S is the spherical distance between the geographic location A and the geographic location B, β1 is the latitude of the geographic location A, α1 is the longitude of the geographic location A, β2 is the latitude of the geographic location B, and α2 is Longitude of location B.
The method according to claim 3, wherein determining the optimal geographic location corresponding to each IP by using the optimization algorithm according to the geographical location clustering result comprises:

For each geographic location in the high density cluster, the weight of each geographic location is determined according to the spherical distance of each geographic location and the high density cluster centroid;

Based on the weights, the optimal geographic location corresponding to each IP is determined using a weighted least squares method.
The method according to claim 5, wherein the weight of each of the geographical locations is determined according to the following formula (2):

Where λ i represents the weight of the i-th geographic location, d i represents the spherical distance between the i-th geographic location and the high-density cluster centroid, and n is an integer greater than or equal to 1;

Determining the optimal geographic location corresponding to the IP according to the following formula (3):

Where (x i , y i ) represents the ith geographic location,
Indicates the optimal geographical location.
The method according to claim 1, wherein determining the precise geographical location of the IP according to the optimal geographic location and a preset artificial neural network model comprises:

Inputting the optimal geographic location into the preset artificial neural network model to obtain an output result;

If the output result is a preset target result, the optimal geographic location is an exact geographic location of the IP.
The method according to claim 7, wherein the input layer of the preset artificial neural network model has three neuron nodes, the hidden layer has five neuron nodes, and the output layer has one neuron node. .
A device for determining a precise geographic location, comprising:

An obtaining module, configured to obtain an IP and multiple geographic locations associated with the IP;

a clustering module, configured to cluster the plurality of geographic locations by using a clustering algorithm to obtain a geographical location clustering result of the IP;

An optimal geographic location determining module, configured to determine, according to the geographic location clustering result, an optimal geographic location corresponding to the IP by using an optimization algorithm;

And a precise geographic location determining module, configured to determine an accurate geographic location of the IP according to the optimal geographic location and a preset artificial neural network model.
The apparatus according to claim 9, wherein said clustering algorithm is a k-means algorithm, and said optimization algorithm is a weighted least squares method.
The device according to claim 10, wherein the clustering module is further configured to:

Selecting two geographic locations from the plurality of geographic locations associated with the IP as the first initial centroid and the second initial centroid;

Calculating a first spherical distance between each of the plurality of geographic locations and the first initial centroid and a second spherical distance from the second initial centroid;

And clustering a plurality of geographical locations associated with the IP to obtain a high density cluster, and using the high density cluster as a geographical location clustering result of the IP according to the first spherical distance and the second spherical distance .
The apparatus according to claim 11, wherein said clustering module calculates a first spherical distance between each geographic location and said first initial centroid and a second initial centroid according to formula (1) below The second spherical distance between:

S=R·ar cos(cosβ1·cosβ2·cos(α1-α2)+sinβ1·sinβ2) (1)

Where R is the radius of the long axis of the earth, S is the spherical distance between the geographic location A and the geographic location B, β1 is the latitude of the geographic location A, α1 is the longitude of the geographic location A, β2 is the latitude of the geographic location B, and α2 is Longitude of location B.
The device according to claim 10, wherein the optimal geographic location determining module is further configured to:

For each geographic location in the high density cluster, the weight of each geographic location is determined according to the spherical distance of each geographic location and the high density cluster centroid;

Based on the weights, the optimal geographic location corresponding to each IP is determined using a weighted least squares method.
The apparatus according to claim 13, wherein the weight of each of the geographical locations is determined according to the following formula (2):

Where λ i represents the weight of the i-th geographic location, d i represents the spherical distance between the i-th geographic location and the high-density cluster centroid, and n is an integer greater than or equal to 1;

Determining the optimal geographic location corresponding to the IP according to the following formula (3):

Where (x i , y i ) represents the ith geographic location,
Indicates the optimal geographical location.
The device according to claim 8, wherein the precise geographic location determining module is further configured to:

Inputting the optimal geographic location into the preset artificial neural network model to obtain an output result;

If the output result is a preset target result, the optimal geographic location is an exact geographic location of the IP.
The apparatus according to claim 15, wherein the input layer of the preset artificial neural network model has three neuron nodes, the hidden layer has five neuron nodes, and the output layer has one neuron node. .
An electronic device, comprising:

One or more processors;

a storage device for storing one or more programs,

The one or more programs are executed by the one or more processors such that the one or more processors implement the method of any of claims 1-8.
A computer readable medium having stored thereon a computer program, wherein the program is executed by a processor to implement the method of any of claims 1-8.