A method, terminal device, and storage medium for identifying a specific location.

Abstract

This invention discloses a method, terminal device, and storage medium for identifying specific locations. It can efficiently and accurately identify specific locations automatically. After acquiring specific data, it marks and classifies it, identifies the types of different elements, and constructs element associations centered on specific individuals. Then, it performs deep extraction on the marked specific data to obtain different types of specific related data, such as specific applications and scripts. Furthermore, it obtains the virtual identity information of specific individuals through specific related data, and analyzes the common internet access locations of specific individuals to obtain the actual address of the specific location. Compared with traditional analysis methods that rely on single dimensions such as base station call detail record analysis and inbound and outbound traffic analysis, which can only locate the base station, this method is more accurate, thus achieving efficient and accurate automatic identification of specific locations.

Description

Technical Field

[0001] This invention relates to the field of location identification technology, and in particular to a method, terminal device and storage medium for identifying a specific location. Background Technology

[0002] Currently, existing technologies for discovering specific locations mainly rely on analyzing call and SMS data from telecom operators. By constructing abnormal communication models, specific base stations are identified, leading to the location of GoIP (Go-Injection Protocol) targets. For discovering specific locations within a local network, methods often rely on aggregating the mobile phone trajectories of specific individuals. However, these methods have significant drawbacks: firstly, their effectiveness is low; secondly, mobile phone trajectories typically only pinpoint base stations, covering a large area and failing to accurately identify specific locations. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to provide a method, terminal device and storage medium for identifying specific locations, which can efficiently and accurately perform automated identification of specific locations.

[0004] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is as follows:

[0005] A method for identifying a specific location, comprising the steps of:

[0006] Acquire specific data, and label and classify the specific data;

[0007] Deep extraction is performed on the specific data that has been labeled and classified to obtain different types of specific related data;

[0008] By analyzing the specific related data based on the common elements between different specific events, virtual identity information of specific individuals can be obtained.

[0009] By verifying the network data corresponding to the virtual identity information of the specific person through the network big data platform resources, a list of Internet WiFi networks corresponding to the virtual identity of the specific person is obtained;

[0010] Based on the list of internet WiFi networks, the actual address associated with the broadband account registered under the same WiFi environment is obtained, and the actual address is marked as a specific location.

[0011] To solve the above-mentioned technical problems, another technical solution adopted by the present invention is as follows:

[0012] A location identification terminal device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of the location identification method described above.

[0013] To solve the above-mentioned technical problems, another technical solution adopted by the present invention is as follows:

[0014] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the location identification method described above.

[0015] The beneficial effects of this invention are as follows: After acquiring specific data, it is marked and classified to identify the types of different specific elements, construct element associations centered on specific individuals, and then perform deep extraction on the marked different specific data to obtain different types of specific related data such as specific applications and scripts. Furthermore, after obtaining the virtual identity information of specific individuals through the specific related data, the common Internet access locations of specific individuals are analyzed to obtain the actual address of specific locations. Compared with traditional single-dimensional analysis methods that rely on base station call detail record analysis and inbound and outbound traffic analysis, which can only locate the base station, this method is more accurate, thereby achieving efficient and accurate automatic identification of specific locations. Attached Figure Description

[0016] Figure 1 This is a flowchart illustrating the steps of a method for identifying a specific location according to an embodiment of the present invention;

[0017] Figure 2 This is a schematic diagram illustrating the steps of a method for identifying a specific location in an embodiment of the present invention;

[0018] Figure 3 This is a schematic diagram of another step in a method for identifying a specific location according to an embodiment of the present invention;

[0019] Figure 4 This is a schematic diagram of serial-parallel analysis based on APK file features in a method for identifying a specific location according to an embodiment of the present invention;

[0020] Figure 5 This is a schematic diagram of cluster analysis based on the voiceprint features of a specific person in a method for identifying a specific location according to an embodiment of the present invention.

[0021] Figure 6 This is a schematic diagram illustrating the location discovery method for identifying a specific location according to an embodiment of the present invention.

[0022] Figure 7 This is a schematic diagram of the structure of a location-specific identification terminal device according to an embodiment of the present invention. Detailed Implementation

[0023] To explain in detail the technical content, objectives, and effects of the present invention, the following description is provided in conjunction with the embodiments and accompanying drawings.

[0024] Please refer to Figure 1 A method for identifying a specific location, comprising the steps of:

[0025] Acquire specific data, and label and classify the specific data;

[0026] Deep extraction is performed on the specific data that has been labeled and classified to obtain different types of specific related data;

[0027] By analyzing the specific related data based on the common elements between different specific events, virtual identity information of specific individuals can be obtained.

[0028] By verifying the network data corresponding to the virtual identity information of the specific person through the network big data platform resources, a list of Internet WiFi networks corresponding to the virtual identity of the specific person is obtained;

[0029] Based on the list of internet WiFi networks, the actual address associated with the broadband account registered under the same WiFi environment is obtained, and the actual address is marked as a specific location.

[0030] As described above, the beneficial effects of this invention are as follows: by acquiring specific data and marking and classifying it, different types of specific elements are identified, and element associations centered on specific individuals are constructed. Then, the different specific data that have been marked are deeply extracted to obtain different types of specific related data such as applications and scripts. Furthermore, after obtaining the virtual identity information of specific individuals through the specific related data, the common internet access locations of specific individuals are analyzed to obtain the actual address of specific locations. Compared with traditional single-dimensional analysis methods that rely on base station call detail record analysis and inbound and outbound traffic analysis, which can only locate the base station, this method is more accurate, thereby achieving efficient and accurate automatic identification of specific applications at specific locations.

[0031] Furthermore, after marking the actual address as a specific location, the process also includes:

[0032] Obtain the real-name information of other virtual identities under the same WiFi environment;

[0033] Calculate the real-name registration rate of all virtual identities under the same WiFi environment;

[0034] Different specific locations are classified according to the real-name registration rate.

[0035] As described above, the system obtains the real-name information of all other virtual identities based on the same WiFi environment, calculates the real-name rate of the WiFi environment based on all virtual identities, and if the real-name rate is low, the WiFi environment is more likely to be a specific location, and it is then classified according to the corresponding probability.

[0036] Furthermore, after marking the actual address as a specific location, the process also includes:

[0037] Obtain the WiFi environment corresponding to the specific location, and obtain the mobile phone numbers associated with the same WiFi environment and the list of installed applications;

[0038] The application list is analyzed and compared with specific applications that have already been identified, and the correspondence rate of specific applications is calculated.

[0039] Different specific locations are classified according to the corresponding rate of the specific application.

[0040] As described above, by obtaining all mobile phone numbers and the list of applications installed on the same WiFi environment, and analyzing and comparing the list of applications on all mobile phones with the specific applications already known, the more specific applications on the mobile phone, the higher the probability that the WiFi environment is a specific location. Based on the corresponding probability, the WiFi environment is classified and processed. By using data analysis from multiple dimensions such as the common places where mobile phone numbers access the Internet, the real-name registration rate of the numbers, and the application installation status, the specific location can be more accurately located.

[0041] Furthermore, the specific data includes at least one of the following: application installation package file, QR code image, URL address, chat voice file, and chat context content;

[0042] The deep extraction of the specific data that has been labeled and classified yields different types of specific related data, including:

[0043] Deep extraction is performed on at least one type of data in the specific data to obtain different types of specific related data.

[0044] As described above, by performing deep extraction on at least one of the following data from specific data: application installation package files, QR code images, URL addresses, chat voice files, and chat context content, it is possible to comprehensively extract and analyze different data, thereby improving the correlation between multiple dimensions of the data and the accuracy of the data.

[0045] Furthermore, deep extraction is performed on at least one type of data in the specific data to obtain different types of specific related data, including:

[0046] If the specific data includes an application installation package file, then the hash value of the application installation package file is calculated using a digest algorithm;

[0047] Obtain the application information description file from the application installation package file;

[0048] Extract the characteristic content of the application information description file to obtain specific application information.

[0049] As described above, by performing deep extraction on the application installation package file, the hash value corresponding to the application and the feature content in the application information description file are obtained, thus obtaining specific application information. Based on the specific application information, the association between different specific events can be compared, and specific events using the same specific application can be identified.

[0050] Furthermore, performing deep extraction on at least one type of data from the specific data to obtain different types of specific related data also includes:

[0051] Run the application installation package file using the emulator;

[0052] By deploying network packet capture tools, we can obtain the system links and domain information corresponding to the application installation package files accessed by specific applications through the unified resource location tool.

[0053] As described above, by simulating the running of the application installation package file, the system link and domain name information corresponding to the application installation package file can be obtained after the application installation package file is run.

[0054] Furthermore, deep extraction is performed on at least one type of data in the specific data to obtain different types of specific related data, including:

[0055] If the specific data includes chat voice files and chat context content;

[0056] Cluster analysis was performed on the chat voice files from different specific events to obtain the same human biometric characteristics;

[0057] Text similarity analysis is performed on the chat context content in different specific events to obtain the dialogue script;

[0058] The analysis of specific related data based on common elements between different specific events to obtain virtual identity information of specific individuals includes:

[0059] By associating the biometric characteristics of individuals with their scripts in different specific events, virtual identity information of specific individuals can be obtained.

[0060] As described above, by obtaining a person's biometrics and dialogue scripts from chat voice files and chat context content in specific data, and then by comparing the biometrics and dialogue scripts of people in different specific events, different specific events can be associated, thereby locking down the virtual identity information of a specific person based on the information contained in different specific events.

[0061] Furthermore, the step of verifying the network data corresponding to the virtual identity information of the specific person through network big data platform resources to obtain the list of Wi-Fi networks corresponding to the virtual identity of the specific person includes:

[0062] The virtual identity of the specific person is compared with the broadband real-name account opening information, broadband internet access records and router access records in the network big data platform resources to obtain the list of internet WiFi networks corresponding to the virtual identity of the specific person.

[0063] As described above, by comparing a specific person's virtual identity with their broadband real-name account opening information, broadband internet access records, and router access records, it is possible to quickly identify the list of Wi-Fi networks corresponding to that specific person's virtual identity.

[0064] In another embodiment of the present invention, a location identification terminal device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the various steps of the location identification method described above.

[0065] In another embodiment of the present invention, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps of the location identification method described above.

[0066] The above-described method, terminal device, and storage medium for identifying specific locations in the present invention can be applied to the identification of specific locations on a network, enabling rapid identification of specific locations.

[0067] Example 1

[0068] Please refer to Figure 1 and Figure 2 A method for identifying a specific location, comprising the steps of:

[0069] S1. Acquire specific data and label and classify the specific data; the specific data is mobile phone data, which is obtained by collecting point-to-point chat records between specific individuals, such as chat context, chat voice files, APP installation package files (Android application package, APK), website URLs or QR codes, etc.; further, the collected data is customized and labeled to establish a strong correlation between the collected data and specific events; the key elements in the collected data are classified to distinguish virtual identities, mobile phone numbers, APPs, URLs, transfer QR codes, APP download QR codes, chat dialogue contexts, chat voices, etc., and to clarify the extraction methods of different data.

[0070] S2. Perform deep extraction on the specific data that has been marked and classified to obtain different types of specific related data; perform deep extraction on at least one of the application installation package file, QR code image, URL address, chat voice file and chat text content to obtain different types of specific related data;

[0071] If the specific data includes application installation package files, then a combination of static and dynamic parsing methods will be used to comprehensively extract information from the specific application installation package files; taking APK files as an example:

[0072] Static analysis: The hash value of the application installation package file is calculated using a hash algorithm. The application information description file (AndroidManifest.xml) within the package is then obtained. Feature content from this file is extracted to obtain specific application information. For example, using the MD5 hash algorithm to calculate the MD5 hash value of the original APK file, and then reading the AndroidManifest.xml file (the application's information description file) from the APK, extracting feature content from specific areas can reveal information such as the app icon, package name, and signature information.

[0073] Dynamic parsing: Run the application installation package file through an emulator; obtain the system link and domain information corresponding to the background Uniform Resource Locator (URL) of the application installation package file accessed by the specific application by deploying a network packet capture tool; for example, run the APK file using an Android emulator, and obtain the background URL (Uniform Resource Locator) link, domain name and other information accessed by the APP by deploying a network packet capture tool (such as Wireshark, HttpWatch, etc.);

[0074] If the specific data includes chat voice files, then cluster analysis is performed on the chat voice files in different specific events to obtain the same human biometric features; the chat voice files are then subjected to data governance work such as quality inspection, preprocessing, effective sound extraction, noise reduction and enhancement, and human voice separation using a voiceprint engine to generate voiceprint information corresponding to each specific person, and the constructed voiceprint data is grouped to form human-centered voiceprint grouping result data;

[0075] If the specific data includes chat context content, then text similarity analysis is performed on the chat context content in different specific events to obtain the dialogue script; the corresponding chat content text is quickly extracted by the type of tag.

[0076] If the specific data includes a QR code image, then information such as the APP download link, website address, and WeChat or Alipay payment account can be obtained by recognizing the QR code image;

[0077] If the specific data includes a URL address, then information such as the domain name, IP address, and URL characteristics of the URL are obtained through regular expressions;

[0078] S3. Analyze the specific related data based on the common elements between different specific events to obtain the virtual identity information of specific personnel; associate the same tools or means, personnel biometrics and scripts used in different specific events to obtain the virtual identity information of specific personnel.

[0079] S4. Verify the network data corresponding to the virtual identity information of the specific person through the network big data platform resources to obtain a list of WiFi networks corresponding to the virtual identity of the specific person; compare the virtual identity of the specific person with the broadband real-name account opening information, broadband internet access records and router access records in the network big data platform resources to obtain a list of WiFi networks corresponding to the virtual identity of the specific person; the broadband internet access records include WiFi connection data of virtual identity and mobile phone number, etc.

[0080] S5. Obtain the actual address associated with the broadband account registered under the same WiFi environment based on the WiFi list, and mark the actual address as a specific location;

[0081] In an optional implementation, to further verify the accuracy of the location-specific analysis, the analysis of the location is performed by using the real-name information of other virtual identities under the same Wi-Fi environment, including:

[0082] S51a, Obtain the real-name information of other virtual identities under the same WiFi environment;

[0083] S52a. Calculate the real-name rate of all virtual identities under the same WiFi environment;

[0084] S53a. Classify the specific locations according to the real-name registration rate;

[0085] Alternatively, analysis can be performed by linking phone numbers connected to the same Wi-Fi network and listing the apps installed, including:

[0086] S51b: Obtain the WiFi environment corresponding to the specific location, and obtain the mobile phone numbers associated with the Internet access in the same WiFi environment and the list of installed applications;

[0087] S52b: Analyze and compare the application list with the specific applications already known, and calculate the correspondence rate of the specific applications; wherein, scheme a and scheme b can also be used simultaneously to improve the accuracy of the analysis of specific locations.

[0088] Example 2

[0089] This implementation demonstrates the application of a location-specific identification method in a real-world scenario.

[0090] Please refer to Figure 3 S1. Obtain the specific data through personnel related to a specific event; the specific data includes: point-to-point chat logs, voice recordings, APP installation package files, URLs or QR codes, etc., between specific personnel. Then, the collected relevant data is customized and labeled. Clear labeling information can establish a strong correlation between the collected data and the specific event. The collected data is grouped according to specific event types, and specific events corresponding to online instant messaging tools such as QQ and WeChat are selected as data sources. The key elements in the collected data are classified to distinguish virtual identities, mobile phone numbers, APPs, URLs, transfer QR codes, APP download QR codes, chat dialogue context, chat voice, etc., and to clarify the extraction methods of different data.

[0091] S2. Perform deep extraction on the specific data that has been marked and classified, and extract APK files, QR code images, URL addresses, chat voice files, chat context content, QQ / WeChat or other virtual identities of specific individuals, mobile phone numbers, etc.

[0092] S3. If it is not possible to directly obtain the QQ / WeChat or other virtual identity of a specific person, then analyze the specific related data based on the common elements between different specific events;

[0093] Please refer to Figure 4 By utilizing the data obtained through data classification and deep extraction, and performing serial-parallel analysis based on the same elements, we can obtain the virtual identity information of specific individuals associated with each element. The more information associated with each identical element, the greater the serial-parallel weight and the stronger the common features. Taking APK file serial-parallel analysis as an example, by analyzing four types of elements—APK package name, signature MD5, icon logo MD5, and main website address—we can find the same items in different APK files. When any one or more elements are the same in different APKs, it means that the two apps have common features, that is, they can be considered as the same application with different names. At this time, we can extract the specific events and virtual identity account information of the corresponding APK. The APP download link, website address, and payment account extracted from the QR code can also be used as features for serial-parallel analysis to discover the virtual identities corresponding to common specific events.

[0094] Please refer to Figure 5 If specific data includes voice files, then the voice files are used as relevant content for specific events. Further clustering analysis is performed using the extracted voiceprint grouping results of specific individuals to find specific events associated with the same voiceprint features and the corresponding virtual identity information of specific individuals.

[0095] Meanwhile, the same specific events often have unified scripts, techniques and routines; by analyzing the chat content between specific people, SimHash calculation is performed on each context to obtain the result (the entire hash calculation involves text segmentation, hash calculation, weighting, merging, dimensionality reduction and other operations), and by analyzing the Hamming distance of the hash values ​​of the chat content between different specific people, the text similarity can be compared to discover scripts with common characteristics.

[0096] Specifically: For example, text1 represents the chat context between specific person A and related person X, and text2 represents the chat context between specific person B and related person Y. Calculate the SimHash results for each of these two specific events. Using hashValue1 = SimHash(text1) and hashValue2 = SimHash(text2), obtain the SimHash results for the two chat contexts respectively. Then, use the following algorithm to determine the maximum bit length of both:

[0097] hashBit=Max(len(bin(hashValue1.value)),len(bin(hashValue2.value),

[0098] At this point, the Hamming distance value is: distance = hashValue1.distance(hashValue2), and the similarity between the two texts is: similar = distance / hashBit. The higher the similarity, the greater the likelihood that the two specific events are related. Through this method, we can obtain a batch of virtual identity information of specific individuals with high similarity.

[0099] S4. Based on the virtual identity information of specific individuals obtained in the above steps, combined with the resources of the network big data platform, and relying on the fixed network monitoring data such as broadband real-name account opening information, broadband internet access records, and router access records held by the big data platform, the list of WIFIs where these virtual identities access the internet can be quickly found.

[0100] S5. Further, find the actual addresses associated with the broadband accounts registered under the same WIFI environment, and mark these addresses as specific locations;

[0101] Please refer to Figure 6To further verify the accuracy of the location-specific analysis, the real-name information of other virtual identities under the same Wi-Fi environment was used to calculate the real-name rate of virtual identities. The lower the real-name rate, the more virtual identities were used without real-name registration, which means the higher the probability of a location-specific location. At the same time, the mobile phone numbers connected to the same Wi-Fi network and their list of installed apps were associated with the analysis. The more apps that were known to be in the app list, the higher the probability of a location-specific location.

[0102] Example 3

[0103] A location identification terminal device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the various steps of a location identification method as described in Embodiment 1 or 2.

[0104] Example 4

[0105] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of a location identification method as described in Embodiment 1 or 2.

[0106] In summary, the method, terminal device, and storage medium for identifying specific locations provided by this invention extract mobile phone information of individuals related to specific events point-to-point and label different element types, constructing element associations centered on specific individuals, which provides important support for subsequent analysis. Furthermore, by deeply extracting different specific data, different types of specific related data, such as specific applications and scripts, are obtained. After further obtaining the virtual identity information of specific individuals, the common internet access locations of mobile phone numbers are analyzed to obtain the actual address of the specific location. The specific location is verified using data analysis from multiple dimensions, such as common internet access locations of mobile phone numbers, number real-name registration rate, and APP installation status. Compared with traditional analysis methods that rely only on single dimensions such as base station call detail record analysis and inbound / outbound traffic analysis, this method is more accurate. The entire analysis process is fully automated, realizing a closed loop of business operations from the occurrence of a specific event, mobile phone data collection and labeling, target data classification, deep feature extraction and feature commonality analysis, common internet access location analysis, specific location discovery, and personnel location tracking, thereby achieving efficient and accurate automated identification of specific locations.

[0107] The above description is merely an embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent modifications made based on the content of the present invention specification and drawings, or direct or indirect applications in related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A method for identifying a specific location, characterized in that, Including the following steps: Acquire specific data, and label and classify the specific data, wherein the specific data includes at least one of the following: application installation package file, QR code image, URL address, chat voice file, and chat context content; Deep extraction is performed on the specific data that has been labeled and classified to obtain different types of specific related data; Based on the common elements between different specific events, the specific related data are analyzed to obtain the virtual identity information of specific personnel. Specifically, this includes associating the biometric characteristics and scripts of personnel in different specific events to obtain the virtual identity information of specific personnel. By verifying the network data corresponding to the virtual identity information of the specific person through the network big data platform resources, a list of Internet WiFi networks corresponding to the virtual identity of the specific person is obtained; Based on the list of internet WiFi networks, the actual address associated with the broadband account registered under the same WiFi environment is obtained, and the actual address is marked as a specific location.

2. The method for identifying a specific location according to claim 1, characterized in that, After marking the actual address as a specific location, the process also includes: Obtain the real-name information of other virtual identities under the same WiFi environment; Calculate the real-name registration rate of all virtual identities under the same WiFi environment; Different specific locations are classified according to the real-name registration rate.

3. The method for identifying a specific location according to claim 1, characterized in that, After marking the actual address as a specific location, the process also includes: Obtain the WiFi environment corresponding to the specific location, and obtain the mobile phone numbers associated with the same WiFi environment and the list of installed applications; The application list is analyzed and compared with specific applications that have already been identified, and the correspondence rate of specific applications is calculated. Different specific locations are classified according to the corresponding rate of the specific application.

4. The method for identifying a specific location according to claim 1, characterized in that, The deep extraction of the specific data that has been labeled and classified yields different types of specific related data, including: Deep extraction is performed on at least one type of data in the specific data to obtain different types of specific related data.

5. The method for identifying a specific location according to claim 4, characterized in that, Deep extraction is performed on at least one type of data from the specific data to obtain different types of specific related data, including: If the specific data includes an application installation package file, then the hash value of the application installation package file is calculated using a digest algorithm; Obtain the application information description file from the application installation package file; Extract the characteristic content of the application information description file to obtain specific application information.

6. A method for identifying a specific location according to claim 4 or 5, characterized in that, Deep extraction of at least one type of data from the specific data to obtain different types of specific related data also includes: Run the application installation package file using the emulator; By deploying network packet capture tools, we can obtain the system links and domain information corresponding to the application installation package files accessed by specific applications through the unified resource location tool.

7. The method for identifying a specific location according to claim 4, characterized in that, Deep extraction is performed on at least one type of data from the specific data to obtain different types of specific related data, including: If the specific data includes chat voice files and chat context content; Cluster analysis was performed on the chat voice files from different specific events to obtain the same human biometric characteristics; Text similarity analysis is performed on the chat context content in different specific events to obtain the dialogue script.

8. The method for identifying a specific location according to claim 1, characterized in that, The process of verifying the network data corresponding to the virtual identity information of the specific person through network big data platform resources to obtain the list of Wi-Fi networks corresponding to the virtual identity of the specific person includes: The virtual identity of the specific person is compared with the broadband real-name account opening information, broadband internet access records and router access records in the network big data platform resources to obtain the list of internet WiFi networks corresponding to the virtual identity of the specific person.

9. A location-specific identification terminal device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements each step of the location identification method as described in any one of claims 1-8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the location identification method as described in any one of claims 1-8.

Images