Methods and related equipment for detecting proxy IP addresses

By using prefix clustering and multi-attribute analysis of known proxy IP addresses, combined with training models and scoring algorithms, proxy IP addresses are proactively detected. This solves the problem of missed detection in existing technologies, improves the accuracy and coverage of detection, and enhances network security.

CN117614653BActive Publication Date: 2026-11-17BEIJING UNIV OF POSTS & TELECOMM
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202311436104.2
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-10-31
Publication Date
2026-11-17
Estimated Expiration
2043-10-31

AI Technical Summary

Technical Problem

In existing technologies, when detecting proxy IP addresses through passive measurement, there is a possibility of missing proxy IP addresses that have not accessed the server, resulting in a small detection coverage and an inability to effectively protect network security.

Method used

By obtaining the prefixes of known proxy IP addresses and clustering them, and combining autonomous system information, domain name information, port attributes, and location attributes, a trained screening model and scoring algorithm are used to actively detect proxy IP addresses and improve detection coverage.

Benefits of technology

It effectively filters out a large number of undetected proxy IP addresses, improving the accuracy and coverage of proxy IP address detection and enhancing network security.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN117614653B_ABST
    Figure CN117614653B_ABST
Patent Text Reader

Abstract

This application provides a method and related equipment for detecting proxy IP addresses. The method includes: acquiring multiple known proxy IP addresses; extracting the prefixes of the known proxy IP addresses and determining the number of known proxy IP addresses with the same prefix; in response to the number of known proxy IP addresses corresponding to any prefix being greater than a preset threshold, acquiring other IP addresses corresponding to that prefix as first predicted IP addresses; determining whether the first predicted IP address belongs to a cloud service proxy based on the autonomous system information and domain name information of the first predicted IP address, thereby obtaining a second predicted IP address; scoring the second predicted IP address based on its port attributes; in response to the score of any second predicted IP address being greater than a preset threshold, determining that second predicted IP address as a target IP address. The method provided by this application can effectively improve the coverage of proxy IP address identification.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This application relates to the field of network security technology, and in particular to a method and related equipment for detecting proxy IP addresses. Background Technology

[0002] A proxy server is a server application that acts as an intermediary between a client requesting resources and the server providing those resources. Its function is to act as a proxy for network users to obtain network information. Figuratively speaking, it's a relay station for network information, a middleman between a personal network and an internet service provider, responsible for forwarding legitimate network information and controlling and registering the forwarding. The proxy IP address (Internet Protocol Address) is the IP address of the proxy server, used to identify the server providing the proxy service. In this mode, the client does not directly connect to the requested target server, but instead directs the request to the proxy server, which then evaluates the request and executes the necessary network tasks. This mechanism simplifies the complexity of requests and is also important for achieving traffic balancing and protecting user privacy.

[0003] For reasons such as network security, saving IP address costs, and improving access speed, proxy servers are sometimes set up between the client and the website server, allowing the client to connect to the website server indirectly through the proxy server. The proxy server acts as an intermediary between the client and the website server, relaying network requests and information. In this case, from the website server's perspective, the network request originates from the proxy server's IP address, i.e., the proxy IP address.

[0004] An anonymous proxy server is a type of server that works through web forms, allowing all internet requests to be filtered through forms first, thus essentially masking the user's identity.

[0005] Anonymous proxy services possess a dual nature in the internet environment. While providing users with personal privacy protection and anonymity, they can also be used by malicious actors to conceal their true identity and geographical location for illicit online activities. This double-edged sword characteristic means that while anonymous proxy services safeguard personal privacy, they also pose cybersecurity vulnerabilities.

[0006] In related technologies, proxy IP addresses are mainly identified by analyzing access traffic or IP login locations. However, these methods typically rely on passive measurement on the server side, which only analyzes proxy IPs that have accessed the server, resulting in limited detection coverage. Summary of the Invention

[0007] In view of this, the purpose of this application is to solve the problems mentioned in the background art and to propose a method and related equipment for detecting proxy IP addresses.

[0008] To achieve the above objectives, this application provides a method for detecting proxy IP addresses, characterized by comprising:

[0009] Obtain multiple known proxy IP addresses, extract the prefixes of the known proxy IP addresses and determine the number of known proxy IP addresses with the same prefix. In response to the number of known proxy IP addresses corresponding to any prefix being greater than a preset threshold, obtain other IP addresses corresponding to that prefix as the first predicted IP address.

[0010] Based on the autonomous system information and domain name information of the first predicted IP address, determine whether the first predicted IP address belongs to a cloud service agent, and obtain the second predicted IP address;

[0011] The second predicted IP address is scored based on its port attribute. If the score of any second predicted IP address is greater than a preset threshold, the second predicted IP address is determined to be the target IP address.

[0012] Optionally, determining whether the first predicted IP address belongs to a cloud service reseller based on the autonomous system information and domain name information of the first predicted IP address includes:

[0013] Obtain the routing information database and the mounted domain name information of the first predicted IP address;

[0014] Based on the routing information database, determine the autonomous system to which the first predicted IP address belongs;

[0015] The allocator is determined based on the Autonomous System ID, and keyword information from the allocator's homepage is obtained;

[0016] Based on the domain name information and the keyword information, a trained screening model is used to determine whether the first predicted IP address belongs to a cloud service agent.

[0017] Optionally, determining whether the first predicted IP address belongs to a cloud service reseller based on the domain name information and the keyword information using a trained filtering model includes:

[0018] Based on the domain name information, determine the average and maximum values ​​of all domain names mounted on the first predicted IP address;

[0019] The keyword information is compared with preset reference keywords to obtain the reference keyword hit value and reference keyword percentage value of the first predicted IP address;

[0020] The average value, maximum value, comparison keyword hit value, and comparison keyword percentage value are input into the trained screening model to determine whether the first predicted IP address belongs to a cloud service agent.

[0021] Optionally, the method for generating the comparison keywords includes:

[0022] Obtain the training allocator's homepage; the allocator is a cloud service agent.

[0023] Based on the allocator's homepage, multiple keywords and their corresponding frequencies are obtained using a trained TF-IDF model;

[0024] The keywords are sorted in a positive order based on frequency.

[0025] Select a preset number of keywords as comparison keywords.

[0026] Optionally, the training method for the screening model includes:

[0027] Obtain the training domain name information, training keyword information, and the tag corresponding to the training IP address;

[0028] A training set is constructed based on the training domain name information, the training keyword information, and the tags;

[0029] Based on the training domain name information, training keyword information, and binary classification model, the training results are obtained;

[0030] The training results and the labels are compared. In response to the accuracy of the training results reaching a preset threshold, the binary classification model is used as the trained screening model.

[0031] Optionally, it also includes:

[0032] The second predicted IP address is scored based on its port and location attributes.

[0033] Optionally, the scoring method includes:

[0034] Based on the port attributes of the second predicted IP address, determine the number of open ports for each second predicted IP address;

[0035] Based on the location attributes of the second predicted IP address, determine the location score for each second predicted IP address;

[0036] Based on the number of ports and the score, a score for each of the second predicted IP addresses is obtained through a preset scoring model.

[0037] Based on the same inventive concept, one or more embodiments of this specification also provide a device for detecting proxy IP addresses, comprising:

[0038] The first prediction module is configured to obtain multiple known proxy IP addresses, extract the prefix of the known proxy IP addresses and determine the number of known proxy IP addresses with the same prefix, and in response to the number of known proxy IP addresses corresponding to any prefix being greater than a preset threshold, obtain other IP addresses corresponding to that prefix as the first predicted IP address.

[0039] The second prediction module is configured to determine whether the first predicted IP address belongs to a cloud service agent based on the autonomous system information and domain name information of the first predicted IP address, and obtain the second predicted IP address.

[0040] The third prediction module is configured to score the second predicted IP address based on the port attribute of the second predicted IP address, and to determine the second predicted IP address as the target IP address in response to the score of any second predicted IP address being greater than a preset threshold.

[0041] Based on the same inventive concept, one or more embodiments of this specification also provide an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the proxy IP address detection method as described in any of the above.

[0042] Based on the same inventive concept, one or more embodiments of this specification also provide a non-transitory computer-readable storage medium storing computer instructions for causing the computer to execute any of the proxy IP address detection methods described above.

[0043] Based on the above, the technical solution of this application, considering the aggregation nature of proxy IPs and their characteristics of autonomous system attributes, domain name attributes, port attributes, and location attributes, firstly clusters all known proxy IP addresses by their prefixes, filtering out known proxy IP addresses with the same prefixes and a large number of them, assuming that other IP addresses under this prefix are also highly likely to be proxy IP addresses; then, based on the autonomous system attributes and domain name attributes of the proxy IP addresses, a trained deep learning model is used to filter the above IP addresses; finally, based on the port attributes and location attributes of the proxy IP addresses, a preset evaluation algorithm is used to perform a final evaluation and filtering of the above IP addresses to obtain the target IP addresses. This application avoids missed proxy IP addresses during the filtering process by actively detecting them, improving the coverage of proxy IP address detection, and further ensuring network security. Attached Figure Description

[0044] To more clearly illustrate the technical solutions in this application or related technologies, the drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the drawings described below are only embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0045] Figure 1 This is a flowchart illustrating a method for detecting proxy IP addresses according to one or more embodiments of this application;

[0046] Figure 2 This is a schematic diagram of the structure of a proxy IP address detection device according to one or more embodiments of this application;

[0047] Figure 3 This is a schematic diagram of the structure of an electronic device according to one or more embodiments of this application. Detailed Implementation

[0048] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with specific embodiments and the accompanying drawings.

[0049] It should be noted that, unless otherwise defined, the technical or scientific terms used in the embodiments of this application should have the ordinary meaning understood by one of ordinary skill in the art to which this application pertains. The terms "first," "second," and similar terms used in the embodiments of this application do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as "comprising" or "including" mean that the element or object preceding the word encompasses the elements or objects listed after the word and their equivalents, without excluding other elements or objects. Terms such as "connected" or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. Terms such as "upper," "lower," "left," and "right" are only used to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0050] As described in the background section, related technologies typically detect proxy IP addresses using a single attribute, such as by detecting proxy IP addresses in data streams. This method usually aggregates data packets into a stream using a 5-tuple (source IP address, destination IP address, source port number, destination port number, and transport layer protocol number), extracts a series of stream features, such as packet size and packet arrival time intervals, and then uses a deep learning model for pattern recognition to detect proxy IP addresses.

[0051] However, the above solution has the following problems: there are obvious missed detections. The above solution mainly analyzes user traffic related to the server through passive measurement, and cannot detect proxy IP addresses that have not accessed the server. That is, only proxy IP addresses that have accessed the server can be detected, and proxy IP addresses that have not accessed the server will not be identified.

[0052] Therefore, this application proposes a method for detecting proxy IP addresses to solve the aforementioned technical problems and improve the accuracy of proxy IP address detection.

[0053] The technical solutions of one or more embodiments of this specification will be described in detail below through specific examples.

[0054] refer to Figure 1 The method for detecting proxy IP addresses according to one embodiment of this specification includes the following steps:

[0055] Step S101: Obtain multiple known proxy IP addresses, extract the prefixes of the known proxy IP addresses and determine the number of known proxy IP addresses with the same prefix. In response to the number of known proxy IP addresses corresponding to any prefix being greater than a preset threshold, obtain other IP addresses corresponding to that prefix as the first predicted IP address.

[0056] In some embodiments, the aforementioned known proxy IP addresses are IP addresses deemed abnormal by data stream detection. In some embodiments, IP addresses deemed abnormal by data stream analysis can be used as known proxy IP addresses, and IP addresses with abnormal login locations can also be used as known proxy IP addresses.

[0057] An IP (Internet Protocol) address consists of two 32-bit parts: a network address and a host address. These parts are divided into four segments, each separated by a period (.). In this application, the first 24 bits of the IP address are referred to as the C segment of the IP address.

[0058] In some embodiments, determining the number of known proxy IP addresses with the same prefix includes determining the number of IP addresses with the same C-class prefix. In some embodiments, the number of bits in the prefix can be adjusted according to the application scenario, for example, set to 16 bits. Different settings of the prefix bit length, as long as they achieve the purpose of this application, are all within the protection scope of this application.

[0059] In the process of implementing this application, the applicant discovered that proxy IP addresses have a clustering tendency. Therefore, in this step, known proxy IP addresses are clustered according to prefixes. When the number of known proxy IP addresses corresponding to a prefix is ​​greater than a preset threshold, it is assumed that there may be other proxy IP addresses among the other IP addresses corresponding to that prefix.

[0060] In some embodiments, the number of known proxy IP addresses corresponding to each prefix can be determined by traversing all known proxy IP addresses and establishing a prefix-quantity dictionary, where the key is the prefix of the known proxy IP address and the value is the number of addresses corresponding to that prefix. That is, traversing all IP addresses to be detected, in response to the fact that the prefix of the known proxy IP address has been included in the above prefix-quantity dictionary, the index value corresponding to the prefix is ​​incremented by one; in response to the fact that the prefix of the known proxy IP address has not been included in the above prefix-quantity dictionary, a new key is established based on the prefix.

[0061] When the number of IP addresses within the same C-class segment of a known proxy IP address exceeds a threshold, based on the clustering of proxy IP addresses, it is reasonable to assume that the IP addresses within that C-class segment are also highly likely to be proxy IP addresses, even though they cannot be detected using current technology. Therefore, this step can filter out a large number of previously undetected proxy IP addresses.

[0062] Step S102: Determine whether the first predicted IP address belongs to a cloud service agent based on the autonomous system information and domain name information of the first predicted IP address, and obtain the second predicted IP address.

[0063] The first predicted IP address has been selected in step S101, but the accuracy of judging proxy IP addresses solely based on clustering is not high. Therefore, further filtering is required in subsequent steps.

[0064] In the process of developing this application, the applicant discovered that IP addresses have multidimensional attributes. IP addresses conform to the standard address format provided by the IP protocol and are used to assign unique logical identifiers to network devices on the Internet. An IP address typically has four key attributes: allocation, port, domain name, and geographical location.

[0065] IP address allocation attributes involve information about the autonomous system to which the IP address belongs and the corresponding organization. The IP address allocation process is carried out by the Internet Corporation for Assigned Numbers (ICANN), which distributes IP addresses to regional Internet Registries. The ICANN then distributes the IP address and its Autonomous System Number (ASN) to Internet Service Providers (ISPs), which ultimately distribute it to the user's host.

[0066] An IP address identifies a host, but a host typically runs multiple applications simultaneously. Therefore, port numbers are used to distinguish data packets from different applications on the same host. The IP address port attribute describes the open ports for a given IP address, which may relate to the availability of a specific application.

[0067] The domain name attribute of an IP address indicates the domain name associated with that IP address, linking the IP address to a domain name. The Domain Name System (DNS) is an internet service. As a distributed database that maps domain names to IP addresses, it makes internet access easier. A DNS hostname is a string of characters separated by dots, representing the name of a computer or group of computers on the internet. It is typically used to name and describe the underlying physical infrastructure that makes up the internet, and may contain information about the device, such as its organization, location, and type of service.

[0068] The geographic attributes of an IP address refer to the geographic location information of the network device to which the given IP address belongs, including information such as the country and city, which provides the basis for realizing geographic location-related functions.

[0069] The following will describe the steps of this application in filtering proxy IP addresses based on the first predicted IP address and the four attributes described above.

[0070] In this step, the initial screening is performed based on the autonomous system attribute and domain name attribute of the IP address to be tested.

[0071] In the process of implementing this application, the applicant discovered that proxy IP addresses are typically assigned by relevant organizations, and by analyzing the keywords on the organization's homepage, it is possible to determine whether the organization is a cloud service reseller. On the other hand, analyzing domain name hosting information also helps in determining whether an organization is a cloud service reseller. Since proxy IP addresses are mostly provided by cloud service resellers, this step primarily determines whether the IP address was assigned by a cloud service reseller based on the autonomous system attributes and domain name attributes of the IP address.

[0072] In some embodiments, determining whether the first predicted IP address belongs to a cloud service reseller based on the autonomous system information and domain name information of the first predicted IP address includes: obtaining the routing information database and the mounted domain name information of the first predicted IP address; determining the autonomous system to which the first predicted IP address belongs based on the routing information database; determining the allocator based on the autonomous system number and obtaining keyword information from the allocator's homepage; and determining whether the first predicted IP address belongs to a cloud service reseller based on the domain name information and the keyword information using a trained filtering model.

[0073] In some embodiments, the domain name information includes mounted domain name information, and the number of mounted domain names for each first predicted IP address and the maximum and average number of mounted domain names for all first predicted IP addresses can be determined based on the domain name information.

[0074] In some embodiments, the aforementioned keyword information may be keywords from an organization's homepage or an organization's email address obtained through web crawling technology. In some embodiments, the aforementioned keywords may be compared with preset comparison keywords, and the comparison results may be used to determine whether the organization is a cloud service agent.

[0075] In some embodiments, determining whether the first predicted IP address belongs to a cloud service agent based on the domain name information and the keyword information using a trained filtering model includes: determining the average and maximum values ​​of the domain names mounted on all the first predicted IP addresses based on the domain name information; comparing the keyword information with preset reference keywords to obtain the reference keyword hit value and reference keyword percentage value of the first predicted IP address; and inputting the average value, maximum value, reference keyword hit value, and reference keyword percentage value into the trained filtering model to determine whether the first predicted IP address belongs to a cloud service agent.

[0076] In some embodiments, the aforementioned keywords can be set manually, for example, based on information such as the name and email address of an existing cloud service agent. In some embodiments, the homepages of a large number of existing cloud service agents can be analyzed to determine the most frequently occurring keywords on the cloud service agent homepages.

[0077] In some embodiments, the method for generating the reference keywords includes: obtaining the homepage of the allocator for training; the allocator is a cloud service agent; obtaining multiple keywords and their corresponding frequencies based on the allocator homepage and a trained TF-IDF model; sorting the multiple keywords in a positive order according to their frequencies; and selecting a preset number of keywords as reference keywords.

[0078] In some embodiments, the training method of the above-mentioned screening model includes: obtaining training domain name information, training keyword information, and a label corresponding to the training IP address; constructing a training set based on the training domain name information, the training keyword information, and the label; obtaining a training result based on the training domain name information, the training keyword information, and the binary classification model; comparing the training result with the label, and in response to the accuracy of the training result reaching a preset threshold, using the binary classification model as the trained screening model.

[0079] In some embodiments, the binary classification model described above can be a random forest model.

[0080] Step S103: Score the second predicted IP address according to the port attribute of the second predicted IP address. If the score of any second predicted IP address is greater than a preset threshold, determine the second predicted IP address as the target IP address.

[0081] Based on step S102, this step further filters the second predicted address.

[0082] In this step, filtering is performed based on the port attribute of the IP address. In some embodiments, filtering can also be performed based on the location attribute.

[0083] In some embodiments, the scoring method includes: determining the number of open ports for each second predicted IP address based on the port attributes of the second predicted IP address; determining the location score for each second predicted IP address based on the location attributes of the second predicted IP address; and obtaining a score for each second predicted IP address based on the number of ports and the score using a preset scoring model.

[0084] In some embodiments, determining the location score of each second predicted IP address based on its location attribute may include: determining the country (or city) of the second predicted IP address, determining the proportion of that country (or city) among all the second predicted IP addresses, and determining the score. For example, if there are five second predicted IP addresses, three located in country A and two located in country B, then when a second predicted IP address is located in country A, its score is 3 / 5.

[0085] In some embodiments, the average and maximum number of open second predicted IP addresses, the average and maximum number of open uncommon ports, and the percentage of second predicted IP addresses with multiple open ports can also be determined based on port attributes.

[0086] In some embodiments, a score can also be calculated based on the number of known proxy IPs corresponding to the C segment of the second predicted IP address.

[0087] In some embodiments, the scoring method includes: manually setting the weight values ​​for the number of ports and the score; multiplying the number of ports by the corresponding weight value to obtain a first multiplication result, and multiplying the score by the corresponding weight value to obtain a second multiplication result; and adding the first multiplication result and the second multiplication result to obtain the score.

[0088] In some embodiments, the scoring method further includes: multiplying the number of known proxy IPs corresponding to the C segment of the second predicted IP address with the corresponding weight value to obtain a third multiplication result, and adding the first multiplication result, the second multiplication result, and the third multiplication result to obtain a score.

[0089] In some embodiments, the scores can be mapped to within 100 according to the autonomous system attribute of the IP address. Specifically, in some embodiments, the autonomous system to which any second predicted IP address belongs is first determined; then, the maximum score among all second predicted IP addresses corresponding to that autonomous system is determined; the score corresponding to the second predicted IP address is divided by the maximum score and then multiplied by 100. This method ensures that the score of the obtained second predicted IP address is within 0-100, facilitating subsequent screening.

[0090] It should be noted that the method in this embodiment can be executed by a single device, such as a computer or server. The method can also be applied in a distributed scenario, where multiple devices cooperate to complete the task. In such a distributed scenario, one of these devices may execute only one or more steps of the method in this embodiment, and the multiple devices will interact with each other to complete the method described.

[0091] It should be noted that the above description describes some embodiments of this application. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in a different order than that shown in the above embodiments and still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0092] Based on the same inventive concept, corresponding to any of the above embodiments, this application also provides a device for detecting proxy IP addresses.

[0093] refer to Figure 2 The device for detecting the proxy IP address includes:

[0094] The first prediction module 11 is configured to obtain multiple known proxy IP addresses, extract the prefix of the known proxy IP addresses and determine the number of known proxy IP addresses with the same prefix, and in response to the number of known proxy IP addresses corresponding to any prefix being greater than a preset threshold, obtain other IP addresses corresponding to that prefix as the first predicted IP address.

[0095] The second prediction module 12 is configured to determine whether the first predicted IP address belongs to a cloud service agent based on the autonomous domain information and domain name information of the first predicted IP address, and obtain the second predicted IP address.

[0096] The third prediction module 13 is configured to score the second predicted IP address based on the port attribute of the second predicted IP address, and to determine the second predicted IP address as the target IP address in response to the score of any second predicted IP address being greater than a preset threshold.

[0097] For ease of description, the above devices are described in terms of function, divided into various modules. Of course, in implementing this application, the functions of each module can be implemented in one or more software and / or hardware.

[0098] The apparatus described above is used to implement the corresponding proxy IP address detection method in any of the foregoing embodiments, and has the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0099] Based on the same inventive concept, corresponding to the methods of any of the above embodiments, this application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the proxy IP address detection method described in any of the above embodiments.

[0100] Figure 3 This embodiment illustrates a more specific hardware structure of an electronic device, which may include a processor 1010, a memory 1020, an input / output interface 1030, a communication interface 1040, and a bus 1050. The processor 1010, memory 1020, input / output interface 1030, and communication interface 1040 are interconnected internally via the bus 1050.

[0101] The processor 1010 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this specification.

[0102] The memory 1020 can be implemented in the form of ROM (Read Only Memory), RAM (Random Access Memory), static storage device, dynamic storage device, etc. The memory 1020 can store the operating system and other applications. When the technical solutions provided in the embodiments of this specification are implemented by software or firmware, the relevant program code is stored in the memory 1020 and is called and executed by the processor 1010.

[0103] The input / output interface 1030 is used to connect input / output modules to realize information input and output. Input / output modules can be configured as components within the device (not shown in the figure) or externally connected to the device to provide corresponding functions. Input devices may include keyboards, mice, touchscreens, microphones, various sensors, etc., while output devices may include displays, speakers, vibrators, indicator lights, etc.

[0104] The communication interface 1040 is used to connect a communication module (not shown in the figure) to enable communication between this device and other devices. The communication module can communicate via wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).

[0105] Bus 1050 includes a pathway for transmitting information between various components of the device, such as processor 1010, memory 1020, input / output interface 1030, and communication interface 1040.

[0106] It should be noted that although the above-described device only shows the processor 1010, memory 1020, input / output interface 1030, communication interface 1040, and bus 1050, in specific implementations, the device may also include other components necessary for normal operation. Furthermore, those skilled in the art will understand that the above-described device may only include the components necessary for implementing the embodiments of this specification, and not necessarily all the components shown in the figures.

[0107] The electronic devices described above are used to implement the corresponding proxy IP address detection method in any of the foregoing embodiments, and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0108] Based on the same inventive concept, corresponding to the methods of any of the above embodiments, this application also provides a non-transitory computer-readable storage medium storing computer instructions for causing the computer to execute the proxy IP address detection method as described in any of the above embodiments.

[0109] The computer-readable medium of this embodiment includes permanent and non-permanent, removable and non-removable media, and information storage can be implemented by any method or technology. Information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic magnetic disk storage or other magnetic storage devices, or any other non-transfer medium that can be used to store information accessible by a computing device.

[0110] The computer instructions stored in the storage medium of the above embodiments are used to cause the computer to execute the proxy IP address detection method as described in any of the above embodiments, and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0111] Those skilled in the art should understand that the discussion of any of the above embodiments is merely exemplary and is not intended to imply that the scope of this application (including the claims) is limited to these examples; within the framework of this application, the technical features of the above embodiments or different embodiments can also be combined, the steps can be implemented in any order, and there are many other variations of different aspects of the embodiments of this application as described above, which are not provided in the details for the sake of brevity.

[0112] Additionally, to simplify the description and discussion, and to avoid obscuring the embodiments of this application, the well-known power / ground connections to integrated circuit (IC) chips and other components may or may not be shown in the provided drawings. Furthermore, the apparatus may be shown in block diagram form to avoid obscuring the embodiments of this application, and this also takes into account the fact that the details of the implementation of these block diagram apparatuses are highly dependent on the platform on which the embodiments of this application will be implemented (i.e., these details should be fully understood by those skilled in the art). While specific details (e.g., circuits) have been set forth to describe exemplary embodiments of this application, it will be apparent to those skilled in the art that the embodiments of this application can be implemented without these specific details or with variations thereof. Therefore, these descriptions should be considered illustrative rather than restrictive.

[0113] Although this application has been described in conjunction with specific embodiments thereof, many substitutions, modifications, and variations of these embodiments will be apparent to those skilled in the art from the foregoing description. For example, other memory architectures (e.g., dynamic RAM (DRAM)) may be used with the embodiments discussed.

[0114] The embodiments of this application are intended to cover all such substitutions, modifications, and variations that fall within the broad scope of the appended claims. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the embodiments of this application should be included within the protection scope of this application.

Claims

1. A method of detecting a proxy IP address, characterized by, include: Obtain multiple known proxy IP addresses, extract the prefixes of the known proxy IP addresses and determine the number of known proxy IP addresses with the same prefix. In response to the number of known proxy IP addresses corresponding to any prefix being greater than a preset threshold, obtain other IP addresses corresponding to that prefix as the first predicted IP address. Based on the autonomous domain information and domain name information of the first predicted IP address, determine whether the first predicted IP address belongs to a cloud service agent, and use the first predicted IP address that belongs to a cloud service agent as the second predicted IP address. The second predicted IP address is scored based on its port attribute. If the score of any second predicted IP address is greater than a preset threshold, the second predicted IP address is determined to be the target IP address. The target IP address is a proxy IP address.

2. The method according to claim 1, characterized in that, The step of determining whether the first predicted IP address belongs to a cloud service reseller based on the autonomous system information and domain name information of the first predicted IP address includes: Obtain the routing information database and the mounted domain name information of the first predicted IP address; Based on the routing information database, determine the autonomous system to which the first predicted IP address belongs; The allocator is determined based on the Autonomous System ID, and keyword information from the allocator's homepage is obtained; Based on the domain name information and the keyword information, a trained screening model is used to determine whether the first predicted IP address belongs to a cloud service agent.

3. The method according to claim 2, characterized in that, The step of determining whether the first predicted IP address belongs to a cloud service reseller based on the domain name information and the keyword information using a trained filtering model includes: Based on the domain name information, determine the average and maximum values ​​of all domain names mounted on the first predicted IP address; The keyword information is compared with preset reference keywords to obtain the reference keyword hit value and reference keyword percentage value of the first predicted IP address; The average value, maximum value, comparison keyword hit value, and comparison keyword percentage value are input into the trained screening model to determine whether the first predicted IP address belongs to a cloud service agent.

4. The method according to claim 3, characterized in that, The method for generating the reference keywords includes: Obtain the training allocator's homepage; the allocator is a cloud service agent. Based on the allocator's homepage, multiple keywords and their corresponding frequencies are obtained using a trained TF-IDF model; The keywords are sorted in a positive order based on frequency. Select a preset number of keywords as comparison keywords.

5. The method according to claim 2, characterized in that, The training methods for the screening model include: Obtain the training domain name information, training keyword information, and the tag corresponding to the training IP address; A training set is constructed based on the training domain name information, the training keyword information, and the tags; Based on the training domain name information, training keyword information, and binary classification model, the training results are obtained; The training results and the labels are compared. In response to the accuracy of the training results reaching a preset threshold, the binary classification model is used as the trained screening model.

6. The method according to claim 1, characterized in that, Also includes: The second predicted IP address is scored based on its port and location attributes.

7. The method according to claim 6, characterized in that, The scoring method includes: Based on the port attributes of the second predicted IP address, determine the number of open ports for each second predicted IP address; Based on the location attributes of the second predicted IP address, determine the location score for each second predicted IP address; Based on the number of ports and the score, a score for each of the second predicted IP addresses is obtained through a preset scoring model.

8. A device for detecting proxy IP addresses, characterized in that, include: The first prediction module is configured to obtain multiple known proxy IP addresses, extract the prefix of the known proxy IP addresses and determine the number of known proxy IP addresses with the same prefix, and in response to the number of known proxy IP addresses corresponding to any prefix being greater than a preset threshold, obtain other IP addresses corresponding to that prefix as the first predicted IP address. The second prediction module is configured to determine whether the first predicted IP address belongs to a cloud service agent based on the autonomous system information and domain name information of the first predicted IP address, and to use the first predicted IP address that belongs to the cloud service agent as the second predicted IP address. The third prediction module is configured to score the second predicted IP address based on the port attribute of the second predicted IP address, and in response to any second predicted IP address having a score greater than a preset threshold, determine that the second predicted IP address is a target IP address; the target IP address is a proxy IP address.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the method as described in any one of claims 1 to 7.

10. A non-transitory computer-readable storage medium storing computer instructions, characterized in that, The computer instructions are used to cause the computer to perform the method according to any one of claims 1 to 7.

Citation Information

Patent Citations

  • Using dhcpv6 and aaa for mobile station prefix delegation and enhanced neighbor discovery

    CN101449529A

  • Method for providing domain name service, system and agent DNS

    CN102333131A