Internet evidence automatic acquisition method and system

Abstract

The invention discloses an internet evidence automatic collection method and system, and the method comprises the steps: collecting the related data of a case, and obtaining the multi-dimensional basic information of the case; the case basic information is deeply integrated, key information points and evidence values thereof are identified, and a case-related original information acquisition strategy is generated according to a set evaluation standard; executing a case-related original information collection strategy, collecting case-related network page contents, and obtaining case-related information; extracting key information by using a natural language processing and image recognition technology, associating the extracted information with a legal knowledge base by using a knowledge graph technology, and generating an optimal evidence fixing strategy based on a rule engine and a machine learning algorithm; and executing the evidence fixing strategy, fixing the evidence involved in the case, generating an evidence collection report, and displaying the evidence collection report through a human-computer interaction interface. According to the invention, the efficiency and accuracy of Internet of Things evidence collection are improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of computer application, and specifically relates to an automatic Internet evidence collection method and system. Background Art

[0002] With the rapid development of Internet technology, the importance of electronic evidence in judicial proceedings has become increasingly prominent. Although there are currently a variety of electronic evidence collection platforms, these platforms mainly provide users with auxiliary tools for fixing evidence. The evidence collection process is essentially still dominated by manual operations. The relevant platforms only play the role of fixing evidence, that is, ensuring that the evidence uploaded to the platform is consistent with the evidence finally downloaded and there is no possibility of modification. However, even if users use such platforms to confirm that the evidence uploaded to the platform is authentic, it is difficult to determine the integrity of the evidence collected and to ensure that a complete chain of evidence has been formed. At the same time, it is also difficult to prove the legality of the entire process of evidence acquisition, as well as key requirements such as the accuracy and relevance of the evidence content. Summary of the invention

[0003] The main purpose of the present invention is to overcome the shortcomings and deficiencies of the prior art and provide a method and system for automatically collecting Internet evidence. By combining OCR technology and knowledge graphs to perform intelligent infringement analysis on the content of web pages, key information is automatically extracted, and the key information is associated with the legal knowledge base to generate an evidence fixation strategy, thereby improving the quality and efficiency of Internet evidence collection.

[0004] In order to achieve the above object, the present invention adopts the following technical solutions: In a first aspect, the present invention provides a method for automatically collecting Internet evidence, comprising the following steps: Collect case-related data and obtain multi-dimensional basic case information; Deeply integrate basic case information, identify key information points and their evidentiary value, and generate strategies for collecting original case information based on set evaluation criteria; Implement the original information collection strategy involved in the case, collect the content of the web pages involved in the case, and obtain the information involved in the case; Extract key information from the case information, associate the key information with the legal knowledge base using the knowledge graph, process the association results based on the rule engine and machine learning, and generate the optimal evidence fixation strategy; the key information includes text information and multimedia information; Execute the evidence fixation strategy, fix the evidence involved in the case, generate an evidence collection report, and display it through the human-computer interaction interface.

[0005] As a preferred technical solution, the deep integration of basic case information includes: classifying the collected basic case information, and comparing the classified information with preset information dimensions; the information dimensions include the type of information and the platform on which the information is located.

[0006] As a preferred technical solution, a historical case library is constructed, including various types of cases and corresponding historical collection strategies; Input the basic information of the case, use the historical collection strategy to identify key information points according to the case type, use the decision tree to classify the key information points, and obtain key information points with evidentiary value; the key information points include infringement behavior, web page location information and electronic media characteristics; According to the set evaluation criteria, key information points with evidentiary value are scored and the scoring results are fed back to the model. If the key information points can be effectively collected, the weight of the historical collection strategy is increased; otherwise, the weight of the historical collection strategy is reduced; the collection strategy is continuously optimized to obtain the original information collection strategy involved in the case.

[0007] As a preferred technical solution, when implementing the strategy for collecting original information involved in the case, if the number of infringing acts increases, an incremental method is used to regularly collect the content of the web pages involved in the case.

[0008] As a preferred technical solution, the strategy for collecting original information involved in the case includes collection objects, collection scope, collection steps, collection tools and collection frequency.

[0009] As a preferred technical solution, the extraction of key information from the case-related information includes: Use optical character recognition to cut the text area of ​​the information involved in the case, identify and extract the text information based on the character characteristics, and correct the text information; Identify objects or signs involved in the case and obtain multimedia information.

[0010] As a preferred technical solution, the method of associating key information with the legal knowledge base using a knowledge graph includes: Encode text information and multimedia information into vector representation to obtain key information vectors; Construct a knowledge graph whose structure includes legal concepts, legal terms and relations; Calculate the similarity between the key information vector and the knowledge graph, and associate the key information and legal terms according to the set threshold.

[0011] As a preferred technical solution, the method of generating the optimal evidence fixation strategy includes the following steps: Rule engine processing: Define the IF-THEN rules of expert knowledge. When key information does not match the legal knowledge base, it is identified as "potential infringement" and the corresponding key information is marked; Machine learning model classification: pre-train machine learning models to classify key information and related results to obtain valid evidence and its importance; Hybrid decision-making: Perform weighted voting on the outputs of rule engine processing and machine learning model classification, determine the evidence collection priority, collection steps, storage format and associated instructions based on the voting results, and obtain the original information collection strategy involved in the case.

[0012] As a preferred technical solution, fixing the evidence involved in the case includes: Taking screenshots, photographs, printing, video recording or notarization of the web pages involved in the case; Upload the evidence involved in the case to the blockchain; Encrypted storage and access control of evidence involved in the case.

[0013] In a second aspect, the present invention further provides an automatic Internet evidence collection system, which is applied to the automatic Internet evidence collection method, and includes an information acquisition module, an information collection scheme generation module, an information collection module, a collection content analysis module, an evidence fixing module and a central control module; Information acquisition module, used to collect case-related data and obtain basic case information; The information collection plan generation module is used to identify key information points and their evidentiary value, and generate the original information collection strategy involved in the case according to the set evaluation criteria; The information collection module is used to implement the original information collection strategy involved in the case, collect the content of the network pages involved in the case, and obtain the information involved in the case; The acquisition content analysis module is used to extract key information from the case information using optical character recognition, associate the key information with the legal knowledge base using the knowledge graph, process the association results based on the rule engine and machine learning, and generate the optimal evidence fixation strategy; the key information includes text information and multimedia information; The evidence fixation module is used to execute the evidence fixation strategy, fix the evidence involved in the case, generate an evidence collection report, and transmit the evidence collection report to the human-computer interaction interface; The central control module is used to coordinate the work of each module, generate evidence collection reports, and operate and view evidence collection reports through the human-computer interaction interface.

[0014] Compared with the prior art, the present invention has the following advantages and beneficial effects: (1) The present invention uses natural language processing technology, image recognition technology and knowledge graph technology to perform intelligent analysis on the collected web page content, which can automatically extract key information and generate the optimal evidence fixation strategy in combination with the legal knowledge base, thereby enhancing the robustness of the model and improving the efficiency and accuracy of evidence collection.

[0015] (2) The present invention processes the association results based on the rule engine and machine learning, which enhances the model's decision-making ability, ensures the scientificity and pertinence of the collection process, reduces labor costs, and improves the quality and efficiency of evidence collection.

[0016] (3) The present invention combines the existing historical case library and provides a set of clear information collection rules, providing a clear and scientific action guide for subsequent practical evidence collection work. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0018] Figure 1 This is a flow chart of a method for automatically collecting Internet evidence according to an embodiment of the present invention; Figure 2 A flowchart generated for the optimal evidence fixation strategy of an embodiment of the present invention; Figure 3 Schematic diagram of the structure of the automatic Internet evidence collection system according to an embodiment of the present invention. DETAILED DESCRIPTION

[0019] In order to enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without making creative work are within the scope of protection of the present application.

[0020] Reference to "embodiments" in this application means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described in this application may be combined with other embodiments.

[0021] Optical Character Recognition (OCR) is a technology that converts text from various printed materials into image information through optical input methods such as scanning, and then uses text recognition technology to convert this image information into usable computer input text.

[0022] Example 1 See also Figure 1 This embodiment provides a method for automatically collecting Internet evidence, comprising the following steps: S1. Collect case-related data and obtain multi-dimensional basic case information; Step S1 is the key starting point of the evidence collection process, which aims to comprehensively collect and analyze basic information related to the case. This includes obtaining relevant information such as the nature of the case, the background of the case, and the specific legal issues involved. It lays a solid foundation for formulating a legal, feasible and efficient evidence collection strategy, and ensures the accuracy and effectiveness of subsequent operations.

[0023] In particular, this embodiment collects case-related information in the following manner: S11. Obtain relevant information of the case on the input interface through dialogue with the user; S12. Obtain case background information from publicly available data sources on the Internet; S13. Other methods that can assist the system in obtaining information.

[0024] It should be noted that the basic case information obtained is multi-dimensional, including case type, name of infringing object, infringing subject and infringing behavior, etc. If necessary, relevant information about the case will also be collected.

[0025] S2. Deeply integrate the basic information of the case, identify key information points and their evidentiary value, and generate a strategy for collecting original information involved in the case based on the set evaluation criteria.

[0026] First, in order to integrate the basic evidence information of the case provided by the user and the clear purpose of rights protection, step S2 deeply integrates the basic information of the case in multiple dimensions, and identifies key information points and the potential evidential value in the information. Among them, the key information points include case-related attribute information, such as the infringement that occurred, the platform where the infringement occurred and the corresponding location, the tools used to determine the subsequent calls and the media used for the infringement, etc.; the potential evidential value in the information is determined around the three characteristics of evidence and the content of proof.

[0027] It is worth explaining that, first, for the deep integration of information, the collected information is classified through a classification algorithm, and then the classified information is compared with the preset information dimensions. The information dimensions here include but are not limited to: ① the type of information, such as video, text, picture and audio, ② the platform where the information is located; second, in the requirements for evidence collection, it is necessary to pay attention to the possibility of evidence being adopted. At this time, it is necessary to evaluate which information is most relevant to the case and which information can form a relatively complete chain of evidence when collecting evidence. This process is to judge the evidentiary value of the information; third, the "three characteristics of evidence" is a relevant expression in the field of law, and its specific meaning is: valid evidence must meet "legality, authenticity, and relevance."

[0028] Secondly, this embodiment will pre-build a historical case library, which contains various types of cases and corresponding collection strategies. The model is trained based on these historical cases, and a preliminary prediction is made to generate a collection strategy.

[0029] Then, based on these preliminary collection plans, intelligent algorithms are used to evaluate each plan from multiple dimensions such as legal effectiveness and technical feasibility, and to derive the optimal strategy for collecting original information involved in the case. The specific content of the collection strategy includes requirements such as "collection objects, collection scope, collection steps, collection tools, and collection frequency", which provides a clear and scientific action guide for subsequent actual evidence collection work.

[0030] Specifically, the evidence collection process is modeled as a decision-making process, and the optimization collection strategy is achieved through interaction with the simulation environment. This embodiment scores key information points with evidentiary value according to the set evaluation criteria, and feeds the scoring results back to the model. When the information collection module fails to effectively collect case information with the same information characteristics, the strategy weight will be reduced. When effective collection can be performed, the strategy weight will be further improved. For example, if an online shopping platform can collect html files in previous cases, the cases on this platform will give priority to collecting html files. The weight adjustment of the collection strategy applies to steps S3-S5. Among them, the legal effect, that is, the probative force of the content that can be collected in such cases, determines the content and steps of the collection, and the technical feasibility is to evaluate the technical characteristics of the target web page and adopt a legal and feasible collection method.

[0031] Finally, the strategy for acquiring the original information involved in the case is passed to step S3, and step S3 executes the relevant strategy for acquiring the original information involved in the case to acquire the information involved in the case. For details, see step S3.

[0032] S3. Implement the strategy for collecting original information involved in the case, collect the content of the web pages involved in the case, and obtain the information involved in the case.

[0033] In step S3, the content of the web pages involved in the case is collected comprehensively, systematically and efficiently, and the integrity, accuracy and legal effect of the collection are ensured through data processing. It should be noted that the information involved in the case collected in step S3 is not the evidence itself, but the collection of relevant information involved in the case before the evidence is determined, which plays the role of big data screening.

[0034] In step S3, when taking the case-related information, the following measures can be taken: When the frequency of infringement increases, incremental data collection will be performed regularly. For example, when a cyber violence case occurs, infringements continue to occur, that is, when the number of people committing cyber violence continues to increase, the model will perform regular incremental data collection to ensure that complete evidence is fixed. Among them, cyber violence is based on the text and image extraction and analysis of the collected information involved in the case, and is matched and judged through a certain sensitive word library. For example, if there is a behavior of wantonly posting insults, slander, and privacy infringement information against individuals on the information network, the sampling frequency of network information will be increased. In addition, this embodiment defines "cyber violence information as: illegal and bad information that is published in the form of text, images, audio, video, etc. to individuals through the Internet, containing insults, slander, incitement to hatred, coercion, invasion of privacy, and accusations, ridicule, denigration, discrimination, etc. that affect physical and mental health. Subsequent matching can confirm whether the corresponding behavior belongs to "cyber violence behavior."

[0035] When collecting case-related information, or after collecting case-related information, this embodiment can display the information collection progress and its collection results in real time and intuitively through a human-computer interaction interface.

[0036] S4. Extract key information from the case information, use the knowledge graph to associate the key information with the legal knowledge base, process the association results based on the rule engine and machine learning, and generate the optimal evidence fixation strategy; the key information includes text information and multimedia information.

[0037] Step S4 will conduct in-depth analysis and intelligent processing of the case information. By integrating advanced artificial intelligence technology and professional legal knowledge graph system, this embodiment can extract key evidence with legal significance from massive information and formulate the final evidence fixation strategy.

[0038] Specifically, the steps include: S41. Use optical character recognition to cut the text area of ​​the information involved in the case, recognize and extract the text information according to the characteristics of the characters, and correct the text information; recognize the objects or signs involved in the case and obtain multimedia information.

[0039] The multimedia information includes images, videos, audio and other audio-visual files. In this embodiment, optical character recognition is used to identify and scan objects, logos or commodities involved in the case on the web page to obtain image information; various operations or actions involving objects, logos or commodities involved in the transaction process are scanned and edited into video files; the recognized audio files are converted into text format, and optical character recognition is used to form images with characters to obtain audio information.

[0040] In addition, optical character recognition can also be used to collect information that can only be captured or recorded.

[0041] S42, please refer to Figure 2 ,Using knowledge graph technology, the extracted information is associated with the legal knowledge base.

[0042] Specifically, text information and multimedia information are encoded into vector representations to obtain key information vectors; a knowledge graph is constructed, whose structure includes legal concepts, legal terms and relationships; the similarity between the key information vector and the knowledge graph is calculated, and the key information and legal terms are associated according to the set threshold.

[0043] In a more specific example, the graph structure example can be represented as follows: Knowledge_Graph = { "Trademark Infringement": { "Relevant Articles": ["Article 57 of the Trademark Law", "Article 58 of the Trademark Law"], "Keywords": ["same trademark", "similar trademark", "unauthorized"] }, "Article 57 of the Trademark Law": "Content": "Using a trademark identical to a registered trademark on the same product without the permission of the trademark registrant...", "Key Concepts": ["Unauthorized", "Same Goods", "Same Trademark"] } }.

[0044] S43. Generate the optimal evidence fixation strategy based on the rule engine and machine learning algorithm, as follows: (1) Rule engine processing: Define the IF-THEN rules of expert knowledge. When key information does not match the legal knowledge base, it is identified as "potential infringement" and the corresponding key information is marked.

[0045] The IF-THEN rules of the expert knowledge are, for example: IF the product description contains a registered trademark name AND the seller is not the trademark owner; THEN mark it as potentially infringing AND record the product link, description and price.

[0046] (2) Machine learning model classification: Pre-train machine learning models to classify key information and related results to obtain valid evidence and its importance.

[0047] Machine learning models use supervised learning methods, such as random forests or support vector machines, to handle complex or ambiguous situations. Specifically, the following steps are taken for classification: Feature engineering: Text features, image features, knowledge graph association results, etc. are used as input features. Model training: The model is trained using annotated historical case data. Prediction: Classify new cases to determine whether they are valid evidence and their importance.

[0048] (3) Hybrid decision-making: The outputs of rule engine processing and machine learning model classification are weighted and voted on. The evidence collection priority, collection steps, storage format and related descriptions are determined based on the voting results to obtain the strategy for collecting the original information involved in the case.

[0049] It needs to be explained that "collection steps" refers to the order in which multiple pieces of evidence are collected, and "association description" exists in the evidence collection report to explain the content to be proved in the case.

[0050] S5. Execute the evidence fixation strategy, fix the evidence involved in the case, generate an evidence collection report, and display it through the human-computer interaction interface.

[0051] In order to accurately implement the evidence fixation strategy, this embodiment will perform the following measures: S51. Screenshot, photograph, print, video record or notarize the web pages involved in the case; S52. Upload the relevant evidence to the blockchain and use blockchain technology to ensure the immutability and traceability of the evidence.

[0052] S53. Encrypt and store the evidence involved in the case and control access to it.

[0053] Example 2 An automatic Internet evidence collection method comprises the following steps: S1. Through the dialogue between the system and the user, the basic information of the case including "the case type is trademark infringement, the name of the infringed trademark, the infringing entity is a certain technology company, and the infringing behavior is that the certain technology company sells goods with the same or similar trademarks on the e-commerce platform", and after the above dialogue, the registration information of the infringed trademark is obtained from the official website of the Trademark Office, and the industrial and commercial registration information of the certain technology company is obtained from the enterprise credit information disclosure system. At the same time, the search engine is used to obtain basic information such as news reports related to Trademark A and the certain technology company.

[0054] S2. The information collection plan generation module will automatically generate a case-related original information collection strategy based on the input information and the collected basic information and the historical case data training model. The specific information collection strategy for this case is as follows: "Determine the target website for evidence collection as various e-commerce platforms (such as Taobao, JD.com, Pinduoduo, etc.), set the corresponding search keywords as the name of a certain technology company and its variants, set the depth of evidence collection to the first 10 pages of each search result, set the frequency of evidence collection to once a day, and collect for 10 consecutive days.

[0055] S3. Execute the generated strategy on multiple e-commerce platforms to collect the product information, store information, user reviews and other data of the instructions, and record metadata such as the collection time and webpage URL.

[0056] S4. To process the collected data, the system first uses OCR technology to identify the trademark logo in the product image, then analyzes the product description through natural language processing technology, extracts the key information, and uses knowledge graph technology to associate the extracted information with the relevant provisions of the Trademark Law, thereby generating the optimal evidence fixation strategy. S5. Execute the optimal evidence fixation strategy generated based on the above analysis results, such as "taking full-page screenshots of the infringing product page, recording a video of browsing product details, and uploading screenshots, videos and other evidence files to blockchain storage" and other operations to fix evidence.

[0057] S6. Integrate all the information and eventually automatically collect a complete chain of evidence, such as "list of infringing goods, detailed information on the machine, list of evidence documents" and other fixed evidence outputs, and automatically generate an evidence collection report to explain the evidence collection process to the user and point out the reasonable and effective use of the evidence collected by the system.

[0058] It should be noted that, for the sake of convenience, the aforementioned method embodiments are all expressed as a series of action combinations, but those skilled in the art should know that the present invention is not limited to the described order of actions, because according to the present invention, certain steps can be performed in other orders or simultaneously.

[0059] Based on the same idea as the automatic Internet evidence collection method in the above embodiment, the present invention also provides an automatic Internet evidence collection system, which can be used to execute the above automatic Internet evidence collection method. For ease of explanation, the structural diagram of the embodiment of the automatic Internet evidence collection system only shows the parts related to the embodiment of the present invention. Those skilled in the art can understand that the illustrated structure does not constitute a limitation on the device, and may include more or fewer components than shown in the diagram, or combine certain components, or arrange the components differently.

[0060] See also Figure 3 In another embodiment of the present application, there is provided an automatic Internet evidence collection system 10, the system comprising an information acquisition module 11, an information collection scheme generation module 12, an information collection module 13, a collection content analysis module 14, an evidence fixing module 15 and a central control module 16; The information acquisition module 11 is used to collect case-related data and obtain basic case information; The information collection scheme generation module 12 is used to identify key information points and their evidentiary value, and to generate a strategy for collecting original information involved in the case according to the set evaluation criteria; The information collection module 13 is used to execute the original information collection strategy involved in the case, collect the content of the network page involved in the case, and obtain the information involved in the case; The acquisition content analysis module 14 is used to extract key information from the case information using optical character recognition, associate the key information with the legal knowledge base using the knowledge graph, process the association results based on the rule engine and machine learning, and generate the optimal evidence fixation strategy; the key information includes text information and multimedia information; The evidence fixation module 15 is used to execute the evidence fixation strategy, fix the evidence involved in the case, generate an evidence collection report, and transmit the evidence collection report to the human-computer interaction interface; The central control module 16 is used to coordinate the work of each module, finally generate an evidence collection report, and operate and view the evidence collection report through a human-computer interaction interface.

[0061] It should be noted that the automatic Internet evidence collection system of the present invention corresponds one-to-one to the automatic Internet evidence collection method of the present invention. The technical features and beneficial effects described in the above-mentioned embodiment of the automatic Internet evidence collection method are applicable to the embodiment of the automatic Internet evidence collection method. For specific contents, please refer to the description in the embodiment of the method of the present invention. It will not be repeated here. This is hereby declared.

[0062] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0063] The above embodiments are preferred implementation modes of the present invention, but the implementation modes of the present invention are not limited to the above embodiments. Any other changes, modifications, substitutions, combinations, and simplifications that do not deviate from the spirit and principles of the present invention should be equivalent replacement methods and are included in the protection scope of the present invention.

Claims

1. A method for automatically collecting Internet evidence, characterized in that: The steps include: Collect case-related data and obtain multi-dimensional basic case information; Deeply integrate basic case information, identify key information points and their evidentiary value, and generate strategies for collecting original case information based on set evaluation criteria; Implement the original information collection strategy involved in the case, collect the content of the web pages involved in the case, and obtain the information involved in the case; Extract key information from the case information, associate the key information with the legal knowledge base using the knowledge graph, process the association results based on the rule engine and machine learning, and generate the optimal evidence fixation strategy; the key information includes text information and multimedia information; Execute the evidence fixation strategy, fix the evidence involved in the case, generate an evidence collection report, and display it through the human-computer interaction interface.

2. The method for automatically collecting Internet evidence according to claim 1 is characterized in that: The in-depth integration of basic case information includes: classifying the collected basic case information, and comparing the classified information with preset information dimensions; the information dimensions include the type of information and the platform where the information is located.

3. The method for automatically collecting Internet evidence according to claim 1 is characterized in that: include: Build a historical case library, including various types of cases and corresponding historical collection strategies; Input the basic information of the case, use the historical collection strategy to identify key information points according to the case type, use the decision tree to classify the key information points, and obtain key information points with evidentiary value; the key information points include infringement behavior, web page location information and electronic media characteristics; According to the set evaluation criteria, key information points with evidentiary value are scored and the scoring results are fed back to the model. If the key information points can be effectively collected, the weight of the historical collection strategy is increased; otherwise, the weight of the historical collection strategy is reduced; the collection strategy is continuously optimized to obtain the original information collection strategy involved in the case.

4. The method for automatically collecting Internet evidence according to claim 3 is characterized in that: When implementing the strategy for collecting original information involved in the case, if the number of infringements increases, an incremental approach will be used to regularly collect the content of the web pages involved.

5. The method for automatically collecting Internet evidence according to claim 1 is characterized in that: The evidence collection strategy involves the original information collection strategy, including the collection object, collection scope, collection steps, collection tools and collection frequency.

6. The method for automatically collecting Internet evidence according to claim 1 is characterized in that: The extraction of key information from the case-related information includes: Use optical character recognition to cut the text area of ​​the information involved in the case, identify and extract the text information based on the character characteristics, and correct the text information; Identify objects or signs involved in the case and obtain multimedia information.

7. The method for automatically collecting Internet evidence according to claim 1 is characterized in that: The use of knowledge graph to associate key information with the legal knowledge base includes: Encode text information and multimedia information into vector representation to obtain key information vectors; Construct a knowledge graph whose structure includes legal concepts, legal terms and relations; Calculate the similarity between the key information vector and the knowledge graph, and associate the key information and legal terms according to the set threshold.

8. The method for automatically collecting Internet evidence according to claim 1 is characterized in that: The method of generating the optimal evidence fixation strategy comprises the following steps: Rule engine processing: Define the IF-THEN rules of expert knowledge. When key information does not match the legal knowledge base, it is identified as "potential infringement" and the corresponding key information is marked; Machine learning model classification: pre-train machine learning models to classify key information and related results to obtain valid evidence and its importance; Hybrid decision-making: The outputs of rule engine processing and machine learning model classification are weighted voted on, and the evidence collection priority, collection steps, storage format and associated instructions are determined based on the voting results.

9. The method for automatically collecting Internet evidence according to claim 1 is characterized in that: The said fixing of evidence involved in the case includes: Taking screenshots, photographs, printing, video recording or notarization of the web pages involved in the case; Upload the evidence involved in the case to the blockchain; Encrypted storage and access control of evidence involved in the case.

10. An automatic Internet evidence collection system, characterized in that: The automatic Internet evidence collection method applied to any one of claims 1 to 9 comprises an information acquisition module, an information collection scheme generation module, an information collection module, a collection content analysis module, an evidence fixing module and a central control module; Information acquisition module, used to collect case-related data and obtain basic case information; The information collection plan generation module is used to identify key information points and their evidentiary value, and generate the original information collection strategy involved in the case according to the set evaluation criteria; The information collection module is used to implement the original information collection strategy involved in the case, collect the content of the network pages involved in the case, and obtain the information involved in the case; The acquisition content analysis module is used to extract key information from the case information using optical character recognition, associate the key information with the legal knowledge base using the knowledge graph, process the association results based on the rule engine and machine learning, and generate the optimal evidence fixation strategy; the key information includes text information and multimedia information; The evidence fixation module is used to execute the evidence fixation strategy, fix the evidence involved in the case, generate an evidence collection report, and transmit the evidence collection report to the human-computer interaction interface; The central control module is used to coordinate the work of each module, generate evidence collection reports, and operate and view evidence collection reports through the human-computer interaction interface.

Images