A judicial evidence data analysis method and system

Abstract

The application provides a judicial evidence data analysis method and system, relates to the field of judicial evidence management and data analysis, and overcomes the deficiency of traditional timestamp dependence by analyzing key information in the evidence and identifying logical contradictions between the evidence, so that the real time sequence of an event can be restored, and the judicial justice is improved.

Description

Technical Field

[0001] This application relates to the fields of judicial evidence management and data analysis, and more specifically, to a method and system for analyzing judicial evidence storage data. Background Technology

[0002] In the field of judicial evidence management, to ensure the seriousness and credibility of evidence, electronic evidence management systems are typically used to uniformly manage and time-stamp various types of evidentiary materials. However, when processing large volumes of data, especially computationally intensive tasks such as video transcoding, existing systems may experience delays in the execution of timestamp services due to resource contention. This delay is not a system failure, but rather a physical characteristic of the system under high load; its magnitude is directly related to the amount of data being processed, exhibiting a non-linear time offset. This can cause the official timestamp sequence generated by the system to differ from the actual order of events, potentially providing misleading information during crucial stages such as court hearings and cross-examination, thus affecting judicial fairness.

[0003] The challenge of existing technologies lies in how to move away from the reliance on official system timestamps and instead re-establish logical connections between evidence by deeply analyzing the content information within the evidence, such as specific behaviors in video footage, time points mentioned in audio conversations, and logical relationships within document text, and identify logical contradictions between different pieces of evidence in order to restore the true timeline and truth of the event.

[0004] To address the aforementioned issues, existing technologies urgently need improvement. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this application provides a judicial evidence data analysis method and system, which has the advantages of being able to identify logical contradictions between evidence by analyzing key information within the evidence, overcoming the shortcomings of traditional timestamp reliance, helping to restore the true timeline of events, and improving judicial fairness.

[0006] This application provides a method for analyzing judicial evidence storage data, where the judicial evidence storage data includes video files, audio files, and / or document files; the method includes:

[0007] Based on the judicial evidence data corresponding to the current case, determine the key information of the inherent logical connection of the event, and establish a mapping relationship between the key information and the judicial evidence data and physical storage address; the key information includes the time, place, people and / or behavior information of the event;

[0008] Based on the characteristics, mapping relationships, and preset rules of the first piece of evidence currently specified by the user, the second piece of evidence is determined; the second piece of evidence is other evidence in the judicial evidence storage data that has a direct logical conflict with the first piece of evidence.

[0009] Generate and display logical contradiction information between the first piece of evidence and the second piece of evidence; the logical contradiction information is a contradiction chain diagram with the first piece of evidence node as the central node and the second piece of evidence node radiating outwards; the first piece of evidence node and the second piece of evidence node in the contradiction chain diagram are all linked to the original evidence content through mapping relationships.

[0010] Furthermore, the characteristic information package of the first evidence includes the time, place, people, and / or behavior information of the event described in the first evidence; the preset rule is the rule of legal logical contradiction.

[0011] Based on the characteristic information, mapping relationship, and preset rules of the first evidence currently specified by the user, determine the second evidence, including:

[0012] Based on the characteristics and pre-defined rules of the first piece of evidence, other evidence that directly conflicts with the first piece of evidence is identified as the second piece of evidence from the key information of the mapping relationship.

[0013] Furthermore, based on the characteristic information, mapping relationship, and preset rules of the first evidence currently specified by the user, the second evidence is determined, including:

[0014] The regional attributes for obtaining key information corresponding to the second piece of evidence;

[0015] Determine the source of direct logical conflict based on regional attributes.

[0016] Furthermore, the regional attributes for obtaining key information corresponding to the second piece of evidence include:

[0017] Physical defect feature analysis is performed on the area where the key information corresponding to the second evidence is located to identify image geometric distortion, uneven lighting and / or background noise caused by the evidence carrier or the collection environment, and the physical defect feature analysis results are obtained.

[0018] Content feature analysis is performed on the area where the key information corresponding to the second piece of evidence is located to identify recognition uncertainties caused by text ambiguity, font abnormalities and / or speech accents, and the content feature analysis results are obtained.

[0019] Based on the analysis results of physical defect characteristics and content characteristics, the causal type of regional attributes is determined and used as the regional attributes.

[0020] Furthermore, based on regional attributes, the sources of direct logical conflicts are determined, including:

[0021] Obtain the cause type of the regional attributes corresponding to each key piece of information involved in the logical conflict;

[0022] Based on the cause type of the regional attributes corresponding to each key piece of information, assess the degree of uncertainty of the impact of each key piece of information on the determination of logical conflict.

[0023] The source of direct logical conflict is determined based on the degree of uncertainty.

[0024] Furthermore, based on the causal type of the regional attributes corresponding to each key piece of information, the degree of uncertainty in the determination of logical conflicts is assessed, including:

[0025] Based on the key information involved in the logical conflict, obtain the quantitative indicators of the regional attribute cause type corresponding to the key information; the quantitative indicators include the image geometric distortion degree indicator, the illumination brightness deviation indicator, the background noise signal-to-noise ratio indicator, the text clarity indicator, the character matching degree indicator and / or the speech feature deviation indicator.

[0026] Based on quantitative indicators and pre-set judicial evidence reliability assessment standards, the quantitative indicators are mapped to corresponding uncertainty impact levels, which serve as the degree of uncertainty impact.

[0027] Furthermore, based on quantitative indicators and pre-set judicial evidence reliability assessment standards, the quantitative indicators are mapped to corresponding uncertainty impact levels, which, as the degree of uncertainty impact, include:

[0028] Obtain current case attribute information; current case attribute information includes case type, nature of evidence, and focus of court examination.

[0029] Based on the case type, the nature of the evidence, and the focus of the court's examination, the weight parameters in the preset judicial evidence reliability assessment standard are adjusted to obtain the adjusted judicial evidence reliability assessment standard.

[0030] Based on the quantitative indicators and the adjusted judicial evidence reliability assessment standards, the quantitative indicators are mapped to the corresponding uncertainty impact levels.

[0031] Furthermore, the weighting parameters in the pre-set judicial evidence reliability assessment criteria will be adjusted, including:

[0032] Based on the current case attribute information, match parameter adjustment rules in the preset rule set, and modify the weight parameters based on the parameter adjustment rules; the preset rule set includes the correspondence between case type, evidence nature, focus of cross-examination and parameter adjustment values.

[0033] Furthermore, based on the current case attribute information, matching parameter adjustment rules are selected from the preset rule set, and weight parameters are modified based on these rules, including:

[0034] Based on the case type, evidence nature, and focus of cross-examination corresponding to the current case attribute information, multiple first rules matching the current case attribute information are queried from the preset rule set;

[0035] Determine the degree of matching between each first rule and the current case attribute information, and weight the parameter adjustment values ​​corresponding to multiple first rules according to the degree of matching to determine the final parameter adjustment value, which is used to adjust the weight parameters in the preset judicial evidence reliability assessment standard.

[0036] Furthermore, this application also proposes a judicial evidence storage data analysis system, wherein the judicial evidence storage data includes video files, audio files, and / or document files; the system includes:

[0037] The mapping establishment module is used to determine key information about the inherent logical relationship of an event based on the judicial evidence data corresponding to the current case, and to establish a mapping relationship between key information and judicial evidence data and physical storage addresses; key information includes the time, place, people and / or behavior information of the event.

[0038] The determination module is used to determine the second evidence based on the characteristic information, mapping relationship and preset rules of the first evidence currently specified by the user; the second evidence is other evidence in the judicial evidence storage data that has a direct logical conflict with the first evidence.

[0039] The generation module is used to generate and display logical contradiction information between the first piece of evidence and the second piece of evidence. The logical contradiction information is a contradiction chain diagram with the first piece of evidence node as the central node and the second piece of evidence node radiating outwards. The first piece of evidence node and the second piece of evidence node in the contradiction chain diagram are all linked to the original evidence content through mapping relationships.

[0040] In summary, the judicial evidence data analysis method and system provided in this application overcomes the shortcomings of traditional timestamp reliance by analyzing key information within the evidence and identifying logical contradictions between evidence. This helps to restore the true timeline of events and improves judicial fairness. Attached Figure Description

[0041] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0042] Figure 1 This is a flowchart illustrating the steps of the judicial evidence storage data analysis method disclosed in the embodiments of the present invention;

[0043] Figure 2 This is a schematic diagram of the judicial evidence preservation data analysis system disclosed in an embodiment of the present invention. Detailed Implementation

[0044] The technical solutions of this application will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of this application, and not all embodiments. The components of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0045] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this application, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0046] Traditional judicial evidence management systems suffer from delays in timestamp service execution due to resource contention when handling computationally intensive tasks such as processing large video files. This delay is not a system failure, but rather a physical characteristic of the system under high load, its magnitude directly related to the amount of data being processed, exhibiting a non-linear time offset. This can cause the official timestamp sequence generated by the system to differ from the actual order of events, potentially providing misleading information during crucial stages such as court hearings and cross-examinations, thus affecting judicial fairness.

[0047] To address the aforementioned challenges, this application attempts to establish an inherent logical connection between pieces of evidence by deeply analyzing key information within judicial evidence data such as videos, audio recordings, and documents—e.g., the time, location, people involved, and their actions. In this way, even if official timestamps are distorted, contradictions can be identified through the logical relationships between the evidence, thereby reconstructing the true chain of events.

[0048] In response, this application proposes a method for analyzing judicial evidence storage data, which includes video files, audio files, and / or document files; such as Figure 1 As shown, the method includes:

[0049] S101, Based on the judicial evidence data corresponding to the current case, determine the key information of the inherent logical connection of the event, and establish a mapping relationship between the key information and the judicial evidence data and physical storage address; the key information includes the time, place, people and / or behavior information of the event.

[0050] S102, determine the second evidence based on the characteristic information, mapping relationship and preset rules of the first evidence currently specified by the user; the second evidence is other evidence in the judicial evidence storage data that has a direct logical conflict with the first evidence;

[0051] S103, Generate and display logical contradiction information between the first piece of evidence and the second piece of evidence; the logical contradiction information is a contradiction chain diagram with the first piece of evidence node as the central node and the second piece of evidence node radiating outwards; the first piece of evidence node and the second piece of evidence node in the contradiction chain diagram are both linked to the original evidence content through mapping relationships.

[0052] This application provides a method for analyzing judicial evidence data, wherein judicial evidence data refers to various electronic evidence materials collected and stored in judicial activities, which can be carried in the form of video files, audio files and / or document files, such as surveillance videos, interrogation recordings, electronic contracts, chat logs, etc., and are mainly used as objects for case analysis and evidence review.

[0053] The core innovation of this application lies in establishing a mapping relationship between the key information of the inherent logical connection of events extracted from judicial evidence storage data and the original data. Based on the first piece of evidence specified by the user and combined with preset rules, the application intelligently identifies and visualizes the second piece of evidence that has a direct logical conflict with the first piece of evidence. This solves the problem that existing judicial evidence storage systems cannot accurately restore the true time sequence of events due to timestamp distortion caused by high load. It achieves the effect of getting rid of dependence on unreliable system timestamps and improving the efficiency and accuracy of evidence analysis.

[0054] Specifically, the proposed solution achieves in-depth analysis of judicial evidence data through a series of logically rigorous steps, overcoming the limitations of traditional reliance on timestamps. First, the system performs in-depth content analysis on the judicial evidence data involved in the current case to identify key information reflecting the inherent logical connections between events. This key information refers to the core elements extracted from the judicial evidence data that reflect the inherent logical connections between events. This can include the time, place, people, and / or behavioral information of the event, such as the actions of people in a video, dates mentioned in audio recordings, and specific events recorded in documents. The main purpose is to move away from reliance on official system timestamps and instead establish logical connections through the evidence content itself. Based on this, the system carefully establishes a mapping relationship between this extracted key information and the original judicial evidence data and its physical storage address. The mapping relationship refers to the association established between key information and original judicial evidence data and their physical storage address. This can be achieved using methods such as database indexes, hash tables, or file metadata links. For example, the extracted key information "January 1, 2023, Zhang San in location A" can be bound to the corresponding video file and its storage path on the server. This is primarily to enable rapid location and retrieval of key information and original evidence content, thereby supporting subsequent logical analysis. This initial step is the foundation of the entire solution. It allows the system to break free from the constraints of external timestamps and instead build a reliable internal association system from the evidence content itself, laying a solid foundation for subsequent logical analysis and rapid retrieval.

[0055] Following this, when a user designates a primary piece of evidence as the starting point for analysis, this primary evidence refers to evidence actively specified by the user during the current case analysis process. Its purpose is to serve as the starting point and center for logical conflict detection, guiding the system to search for and analyze relevant evidence. The system acquires the characteristic information of this primary evidence. Subsequently, the system utilizes previously established mapping relationships and combines them with preset rules to intelligently search through all judicial evidence data. The goal of the search is to identify other evidence that directly conflicts with the primary evidence logically, and to designate these identified pieces of evidence as secondary evidence. Secondary evidence refers to other evidence in the judicial evidence data that, after system analysis, is found to have a direct logical conflict with the primary evidence. Its purpose is to identify evidence that contradicts the evidence of interest to the user, providing clues to reveal the truth of the event. The core of this process is that the system no longer simply compares the order of timestamps, but delves into the semantic level of the evidence content. Based on preset logical rules (e.g., the same event cannot occur simultaneously in different locations, or the same person cannot appear in two incompatible scenarios at the same time), it judges whether there are contradictions between the evidence, thereby effectively filtering out evidence that may be questionable.

[0056] Ultimately, once the logical conflict between the first and second pieces of evidence is identified, the system generates and displays this logical contradiction information in an intuitive way. This logical contradiction information refers to the presentation of the logical conflict between the first and second pieces of evidence generated and displayed by the system. Specifically, it is a contradiction chain diagram with the first evidence node as the central node and radiating to connect the second evidence nodes. Its purpose is to present the abstract logical conflict to the user in an intuitive and visual way, facilitating user understanding and analysis. This display format is specifically manifested as a contradiction chain diagram with the first evidence node as the central node and radiating to connect the second evidence nodes. The contradiction chain diagram is a visual structure in which the first evidence node serves as the center, and multiple second evidence nodes radiate to this central node. Furthermore, the first and second evidence nodes in the diagram are linked to the original evidence content through mapping relationships. It can be presented in the form of node-side diagrams, mind maps, or network topology diagrams in a graphical user interface (GUI). Its main purpose is to intuitively display the contradictory relationships between evidence and provide a quick way to access the original evidence, thereby supporting users in conducting in-depth cross-examination and analysis. This graphical presentation allows users to immediately see which pieces of evidence contradict the primary piece of evidence they are focusing on, and the connections between these contradictions. More importantly, each evidence node in the contradiction chain diagram, whether it's the primary or secondary evidence node, is directly linked to its corresponding original evidence content through a previously established mapping relationship. This means users can click on a node at any time to immediately view the original video, audio, or document files, thereby verifying and conducting in-depth analysis of the contradictions. Through this complete workflow, this solution effectively reveals evidentiary contradictions obscured by traditional timestamps, helping analysts reconstruct the true logical chronology of events.

[0057] In some preferred embodiments, it is assumed that in a case, there exists a video file, an audio file, and a document file as judicial evidence data. First, the system will initiate a data parsing module to process this judicial evidence data. For example, for the video file, computer vision technology can be used to identify faces, specific actions, and time displays in the scene; for the audio file, speech recognition technology can be used to convert the spoken content into text, and the time, location, and names of people mentioned in the dialogue can be extracted; for the document file, natural language processing technology can be used to extract the time, location, people, and behavioral descriptions of the events from the text. This extracted information, such as "the video shows Zhang San appearing at location A at 10:00 AM on January 1, 2023," "the audio record shows Li Si mentioning he was at location B at 11:00 AM on January 1, 2023," and "the document records Wang Wu signing a contract located at location C at 10:30 AM on January 1, 2023," will be used as key information. Subsequently, the system will index this key information along with the corresponding original video files, audio files, document files, and their specific storage paths on the server, forming a mapping database that can be quickly queried.

[0058] For example, a user might specify the characteristic information in a video file that "Zhang San appeared in location A at 10:00 AM on January 1, 2023" as primary evidence. Based on this characteristic information, and combined with pre-defined legal logic contradiction rules, such as "a person cannot be in two different locations at the same time," the system will search the mapping database for key information on other evidence. If the system finds that Li Si's dialogue in an audio file mentions "At 10:00 AM on January 1, 2023, I was talking to Zhang San in location A," while a document file records "At 10:00 AM on January 1, 2023, Zhang San signed a document in location B," then the system will identify a direct logical conflict between the relevant content in the audio file and the document file and the primary evidence, and will determine these audio files and document files as secondary evidence.

[0059] Specifically, the system will further generate and display logical contradictions between the first and second pieces of evidence. This will be presented in the form of a contradiction chain diagram, with the central node being "Zhang San appeared at location A at 10:00 AM on January 1, 2023" in the video file, and the radially connected nodes being "Li Si spoke with Zhang San at location A at 10:00 AM on January 1, 2023" in the audio file and "Zhang San signed a document at location B at 10:00 AM on January 1, 2023". Each node will provide a clickable link, allowing users to directly play the original video or audio or open the original document to verify the evidence content. In this way, users can intuitively see the contradictions of Zhang San appearing in different locations at the same time, thereby gaining a deeper understanding of the authenticity of the event.

[0060] Through the aforementioned technical solution, this application effectively addresses the issue of timestamp recording delays caused by high-load tasks in judicial evidence preservation management systems, thereby preventing distortion of the evidence's timeline. By deeply analyzing the inherent logical connections between events within judicial evidence preservation data and establishing a mapping relationship between key information and original evidence, this solution eliminates reliance on external, potentially distorted system timestamps, instead relying on logical judgments based on the evidence content itself. This enables the system to accurately identify direct logical conflicts between primary and secondary evidence, even if the official timestamps of these pieces of evidence deviate. Furthermore, by generating and displaying intuitive contradiction chain diagrams and providing direct links to the original evidence content, this solution significantly improves the efficiency and accuracy of evidence analysis, helping users quickly locate and understand contradictions between pieces of evidence, thus contributing to the reconstruction of the true timeline of events and ensuring judicial fairness.

[0061] This application further proposes that the steps for determining the second piece of evidence include:

[0062] The primary evidence contains information about the time, place, people, and / or actions of the events described in the primary evidence; the presupposed rule is the rule of legal logical contradiction.

[0063] Based on the characteristic information, mapping relationship, and preset rules of the first evidence currently specified by the user, determine the second evidence, including:

[0064] Based on the characteristics and pre-defined rules of the first piece of evidence, other evidence that directly conflicts with the first piece of evidence is identified as the second piece of evidence from the key information of the mapping relationship.

[0065] Among them, the rule of legal logical contradiction refers to the set of rules used in the legal field to determine whether there is a logical inconsistency or mutual exclusion between pieces of evidence. This can include rules of temporal conflict (e.g., the same event cannot occur at two different locations at the same time, or the same subject cannot appear at two different locations at the same time); rules of locational conflict (e.g., the same event cannot simultaneously involve two mutually exclusive actions at the same location); rules of personal conflict (e.g., the same person cannot simultaneously play two contradictory roles); and rules of behavioral conflict (e.g., the same subject cannot simultaneously perform two opposing actions). The purpose is to provide an objective and unified legal basis for judging conflicts of evidence, ensuring the accuracy and legal validity of the judgment. Furthermore, direct logical conflict refers to a clear and irreconcilable logical contradiction between two or more pieces of evidence, making it impossible for them to be true simultaneously. This can manifest as a direct opposition in information about time, place, person, or behavior. For example, one piece of evidence indicates that a person was at a certain time and place, while another piece of evidence indicates that the person was at a different location at the same time. The purpose is to focus on the core point of contradiction, avoid introducing vague or indirect connections, and improve the efficiency and accuracy of conflict identification.

[0066] In some preferred embodiments, suppose a user designates a recorded phone call as primary evidence, mentioning, "I met Zhang San at a downtown coffee shop at 10:00 AM on October 26, 2023." The system then extracts key information from this primary evidence: the event occurred at "10:00 AM on October 26, 2023," the location was a "downtown coffee shop," the person was "Zhang San," and the action was "meeting." Simultaneously, the system applies pre-defined legal logic contradiction rules, such as the rule that "the same person cannot appear in two different locations at the same time." Subsequently, the system uses this extracted key information and the legal logic contradiction rules to search within a previously established key information database. For example, if the database contains a summary of surveillance video information showing "Li Si appeared at a suburban warehouse at 10:00 AM on October 26, 2023," the system compares the person information and finds it does not match the "Zhang San" in the primary evidence, thus not constituting a direct conflict. However, if the key information database contains a summary of another GPS location record, whose key information shows "Zhang San's mobile phone signal appeared in a suburban warehouse at 10:00 AM on October 26, 2023," the system will identify the person "Zhang San" and the time "10:00 AM on October 26, 2023" as matching the first piece of evidence. However, the location "suburban warehouse" directly contradicts the "city center coffee shop" mentioned in the first piece of evidence. Based on the legal logical contradiction rule that "the same person cannot appear in two different locations at the same time," the system will identify this GPS location record as second evidence that directly conflicts with the call recording. In this way, the system can quickly and accurately locate other evidence that clearly contradicts the specified evidence without requiring manual review of all original evidence.

[0067] Through the above technical solution, this application can clearly define the characteristic information and preset rules of the first evidence, and limit the specific steps for finding the second evidence, thereby improving the efficiency and accuracy of evidence analysis. Specifically, by limiting the characteristic information of the first evidence to the time, place, people, and / or behavior information of the event, the search scope can be narrowed, allowing the system to focus on comparing the core elements of the event. Simultaneously, clarifying the preset rules as legal logical contradiction rules ensures that the identified conflicts are based on legal logic, avoiding subjective judgment and improving the accuracy of the judgment. Furthermore, searching for other evidence that directly conflicts with the first evidence within the key information of the mapping relationship allows the search process to be based on structured and refined key information, rather than a comprehensive comparison of the original evidence content, significantly improving search efficiency and ensuring that the identified conflicts are clear and directly determinate. This effectively solves the problem in practical applications of how to effectively utilize multiple information and rules to quickly and accurately find second evidence that logically conflicts with the first evidence.

[0068] This application further proposes steps for determining the second piece of evidence, including:

[0069] Obtain the regional attributes of the key information corresponding to the second piece of evidence; based on the regional attributes, determine the source of the direct logical conflict.

[0070] Among these, regional attributes refer to the characteristics of key information within a specific area of ​​the evidence. These attributes can include the physical quality of the area, its environmental background, or the clarity of the content. Their purpose is to provide foundational information for subsequent assessment of the nature and cause of logical conflicts. Determining the source of direct logical conflict involves identifying the specific reasons for the logical inconsistency between the first and second pieces of evidence. This could be due to issues with the quality of the evidence itself, interference during the collection process, or other factors. The aim is to more accurately assess the reliability of the evidence.

[0071] This application's solution, based on identifying the second piece of evidence, further determines the source of direct logical conflict by acquiring the regional attributes of the key information corresponding to the second piece of evidence. Specifically, in the previous steps, the system has identified the second piece of evidence that directly conflicts with the first piece of evidence based on the characteristics, mapping relationships, and preset rules of the first piece of evidence. However, merely identifying the conflict is insufficient to support a judicial judgment, as the nature and cause of the conflict are still unclear. Therefore, this solution introduces the analysis of the regional attributes of key information in the second piece of evidence. By deeply analyzing the specific environment and background characteristics of the key information, such as its physical quality or content clarity, a basis can be provided for subsequent diagnosis of the source of conflict. Subsequently, based on these regional attributes, the system can identify the specific cause of the logical conflict, such as physical defects in the evidence itself, environmental interference during the collection process, or other factors. This in-depth analysis enables the system not only to discover contradictions between pieces of evidence but also to reveal the root cause of the contradictions, thereby providing judicial personnel with more comprehensive and reliable evidence evaluation information, effectively avoiding misjudgments or difficulties in evidence acceptance due to the inability to determine the source of conflict, and significantly improving the depth and accuracy of evidence analysis.

[0072] In some preferred embodiments, when the system has identified a second piece of evidence that logically conflicts with the first piece of evidence, in order to analyze the source of the conflict in depth, it first obtains the regional attributes corresponding to the key information in the second piece of evidence. For example, if the second piece of evidence is a video file, where the key information is the action of a specific person in the scene, the system can analyze the attributes of the area in the scene where the person is located, such as the clarity of the area, lighting conditions, or whether there are physical defects such as blurriness or jitter. If the second piece of evidence is an audio file, where the key information is a dialogue, the system can analyze the audio attributes of the time period in which the dialogue takes place, such as whether there is background noise, whether the speech is clear, or whether there is distortion. If the second piece of evidence is a document file, where the key information is a number or signature, the system can analyze the attributes of the area where the number or signature is located, such as whether the handwriting is blurry, whether the font is abnormal, or whether there are scanning defects.

[0073] Furthermore, based on the acquired regional attributes, the system can determine the source of direct logical conflicts. For example, if key information in video evidence is located in an area of ​​insufficient lighting or with obvious geometric distortion, the logical conflict may stem from environmental conditions or equipment defects during video recording. If key information in audio evidence is located in a period of excessive background noise or echo, the conflict may be due to interference from the recording environment. If key information in document evidence is located in a section with illegible handwriting or ink diffusion, the conflict may be due to physical damage to the document itself or scanning quality issues. In this way, the system can link abstract logical conflicts with specific evidentiary quality problems or external interference factors, thereby providing more convincing evidence for judicial decisions.

[0074] Through the aforementioned technical solution, the system, after identifying logical conflicts between pieces of evidence, can further analyze and determine the specific source of those conflicts. This allows judicial personnel to go beyond simply discovering contradictions and understand their root causes, such as the quality of the evidence itself, environmental interference during the collection process, or other factors. Consequently, the reliability and credibility of evidence can be assessed more accurately, providing a more comprehensive and reliable basis for subsequent judicial decisions. This effectively avoids difficulties in evidence acceptance or misjudgments due to the inability to determine the source of conflicts, thus improving the fairness and efficiency of judicial trials.

[0075] This application further proposes steps for obtaining the regional attributes of key information corresponding to the second piece of evidence, including:

[0076] Physical defect feature analysis is performed on the area where the key information corresponding to the second evidence is located to identify image geometric distortion, uneven lighting and / or background noise caused by the evidence carrier or the collection environment, and the physical defect feature analysis results are obtained.

[0077] Content feature analysis is performed on the area where the key information corresponding to the second piece of evidence is located to identify recognition uncertainties caused by text ambiguity, font abnormalities and / or speech accents, and the content feature analysis results are obtained.

[0078] Based on the analysis results of physical defect characteristics and content characteristics, the causal type of regional attributes is determined and used as the regional attributes.

[0079] Physical defect feature analysis refers to detecting the visual or auditory characteristics of the area containing key information to identify and quantify non-content-related distortions introduced by the evidence carrier itself (such as wear and tear, damage) or the acquisition process (such as camera shake, microphone malfunction). This can be achieved using image processing algorithms (such as edge detection, Fourier transform analysis of distortion), photometer measurements (analyzing illumination uniformity), or signal processing techniques (such as spectral analysis to identify noise). Its purpose is to reveal physical problems related to the original quality of the evidence. Content feature analysis refers to parsing the semantic content carried by the key information to assess its readability, recognizability, or comprehensibility. Certainty can be achieved by using Optical Character Recognition (OCR) technology to assess text clarity, Natural Language Processing (NLP) technology to analyze text structure, or Automatic Speech Recognition (ASR) technology to assess speech clarity and accent features. The purpose is to identify and quantify the semantic difficulty or ambiguity of key information. The cause type of regional attributes refers to the classification of the root causes that lead to specific attributes (such as unclear or incomplete) in key information regions. Specifically, it can include "physical damage type", "environmental interference type", "content ambiguity type" or "identification ambiguity type", etc. The purpose is to provide a basis for subsequent judgment of the source of logical conflict and improve the pertinence of the basis.

[0080] This application's solution enhances the accuracy of regional attribute acquisition by analyzing the area containing key information corresponding to the second piece of evidence, thereby increasing the dimensionality and precision of the analysis. Specifically, firstly, through physical defect feature analysis, the system can identify and quantify physical distortions such as image geometric distortion, uneven lighting, and background noise introduced by the evidence carrier or acquisition environment. This allows the system to distinguish regional attribute deviations caused by quality issues with the evidence itself, rather than logical problems in the content. Secondly, through content feature analysis, the system can identify and quantify recognition uncertainties caused by factors such as text blurring, font abnormalities, and speech accents. This allows the system to assess the semantic credibility of key information, avoiding misjudgments due to content recognition errors. Finally, based on the results of these two types of analysis, the system comprehensively determines the causal type of the regional attributes. This comprehensive judgment mechanism can fully consider various factors affecting the accuracy of regional attributes, thereby improving the authenticity and reliability of the regional attributes. This improved accuracy of regional attributes provides a more solid foundation for determining the source of direct logical conflicts based on these attributes. This allows for a deeper analysis beyond the surface, delving into the root causes of the conflict, such as physical damage to the evidence itself or difficulties in content identification. Combined with the aforementioned process of determining second evidence based on the characteristics, mapping relationships, and pre-defined rules of the first piece of evidence, and generating information displaying logical contradictions, the final presented logical contradiction information not only reveals the contradiction itself but also its underlying causes. This enhances the depth and reliability of judicial evidence data analysis, providing support for judicial practice and improving its insightful analysis.

[0081] In some preferred embodiments, specifically when it is necessary to obtain the attributes of a key timestamp region in a video piece of evidence, the system can first perform physical defect feature analysis on the timestamp region. For example, image processing algorithms can be used to detect whether there is geometric distortion in the region, such as image distortion caused by camera shake; simultaneously, the illumination distribution in the region can be analyzed to identify whether there is uneven illumination caused by local overexposure or underexposure; and the presence of background noise can be identified through frequency domain analysis of the image. These analyses will generate a physical defect feature analysis result, such as "slight geometric distortion, uniform illumination, no background noise". Next, the system can perform content feature analysis on the timestamp region. For example, an optical character recognition (OCR) engine can be called to recognize the timestamp number and evaluate its recognition confidence; if the confidence of the recognition result is low, or there are multiple possible recognition results, it can be marked as text blurry or recognition uncertain. This analysis will generate a content feature analysis result, such as "low text recognition confidence, blurry". Finally, the system will comprehensively determine the cause type of the regional attributes of the timestamp region based on these two analysis results. For example, if the physical defect analysis shows no physical defects, but the content feature analysis shows blurry text with low recognition confidence, the system can determine that the cause type of the regional attribute is "content blurry". If the physical defect analysis shows severe uneven lighting, but the content feature analysis shows acceptable text clarity, the system can determine that the cause type is "environmental interference". In this way, the system can provide regional attribute information for subsequent logical conflict determination, improving the detail and accuracy of the regional attribute information.

[0082] This application further proposes steps for determining the source of direct logical conflict based on regional attributes, including:

[0083] Obtain the cause type of the regional attributes corresponding to each key piece of information involved in the logical conflict;

[0084] Based on the cause type of the regional attributes corresponding to each key piece of information, assess the degree of uncertainty of the impact of each key piece of information on the determination of logical conflict.

[0085] The source of direct logical conflict is determined based on the degree of uncertainty.

[0086] The causal type of the regional attributes corresponding to the key information involved in the logical conflict refers to the root cause that leads to the specific attributes (such as blur, noise, distortion) of the key information region. This can be achieved through in-depth analysis of image, audio, or text data. For example, image processing algorithms can be used to identify camera shake, insufficient lighting, or compression artifacts; speech recognition technology can be used to analyze accents or recording environment noise; or text analysis can be used to identify scan quality or original document corruption. The aim is to reveal the true reasons behind the regional attributes, providing a foundation for subsequent reliability assessment. Assessing the degree of uncertainty impact of each key piece of information on the determination of logical conflict refers to quantifying the credibility or uncertainty level of the key information in the determination of logical conflict based on the causal type of the key information regional attributes. This can be achieved by establishing a mapping relationship between different causal types and uncertainty levels. For example, different uncertainty weights can be assigned to blur caused by camera shake and blur caused by original document corruption. Alternatively, expert systems or machine learning models can be used to comprehensively judge the causal types and output quantitative indicators. The aim is to provide a quantitative reliability basis for subsequent determination of the source of conflict. Determining the source of a direct logical conflict refers to identifying the most important or fundamental key information that causes the logical conflict after comprehensively considering the degree of uncertainty of all relevant key information. This can be achieved by comparing the degree of uncertainty of different key information and selecting the key information with the highest or most significant impact as the source of the conflict, or by comprehensively analyzing multiple key information through methods such as weighted voting or Bayesian inference. The purpose is to accurately pinpoint the root cause of the logical conflict and avoid misjudgment.

[0087] In some preferred embodiments, it is assumed that during the analysis of judicial evidence data, the system has identified a logical conflict between the first and second pieces of evidence, and has obtained the regional attributes of key information in the second evidence (e.g., timestamp information in a video and date information in a document). To more accurately determine the source of the conflict, the system first obtains the causal type of the regional attributes corresponding to these key pieces of information. Specifically, for timestamp information in a video, the system may analyze the image quality of its location. If significant geometric distortion or uneven lighting is found, the system will further analyze the causes of these phenomena. For example, image processing algorithms may identify that the distortion is due to wide-angle lens shooting, or that uneven lighting is due to insufficient ambient light. For date information in a document, the system may analyze its text clarity. If blurry text is found, the system will attempt to identify the cause of the blurriness, such as improper scanner settings or ink diffusion or illegible handwriting in the original paper document. Then, based on these causal types, the system will assess the degree of uncertainty affecting the determination of the logical conflict by each piece of key information. For example, if the uncertainty of a video timestamp is assessed as "low," while the uncertainty of a document date is assessed as "high," the system will determine that the document date is more likely to be the direct source of the logical conflict. The system can pre-define a rule base to map different cause types to different uncertainty levels, such as "low," "medium," "high," or specific numerical ranges. The system can use weighted averaging or decision tree models to comprehensively consider the uncertainty of all key information, thereby arriving at the most reasonable conflict source judgment. In this way, the system avoids simply treating all ambiguity or distortion as problems of equal severity, but instead makes refined judgments based on their causes, thus improving the accuracy of conflict source determination.

[0088] Through the above technical solution, this application overcomes the limitations of determining the source of logical conflict solely based on regional attributes. By deeply analyzing the causal types of the regional attributes corresponding to each key piece of information involved in the logical conflict, and assessing the degree of uncertainty in the determination of logical conflict based on these causal types, the source of direct logical conflict can be determined more accurately. This enables the system to effectively distinguish the impact of different types of defects on the reliability of evidence when facing complex and ever-changing judicial evidence data, avoiding misjudgments caused by simplistic judgments, thereby significantly improving the accuracy and reliability of judicial evidence data analysis and providing a more refined and reliable evidence analysis tool for judicial practice.

[0089] This application further proposes steps for assessing the degree of uncertainty regarding the determination of logical conflicts, including:

[0090] Based on the key information involved in the logical conflict, obtain the quantitative indicators of the regional attribute cause type corresponding to the key information; the quantitative indicators include the image geometric distortion degree indicator, the illumination brightness deviation indicator, the background noise signal-to-noise ratio indicator, the text clarity indicator, the character matching degree indicator and / or the speech feature deviation indicator.

[0091] Based on quantitative indicators and pre-set judicial evidence reliability assessment standards, the quantitative indicators are mapped to corresponding uncertainty impact levels, which serve as the degree of uncertainty impact.

[0092] Specifically, this application's solution addresses the problem of inaccurate logical conflict determination due to the inherent quality of evidence in judicial evidence data analysis by introducing a quantitative evaluation mechanism. Specifically, after identifying the second piece of evidence and obtaining the causal types of its key information area attributes, to more accurately determine the source of direct logical conflict, this solution first obtains quantitative indicators of the causal types of the corresponding area attributes for each key piece of information involved in the logical conflict. These quantitative indicators, such as image geometric distortion indicators, illumination deviation indicators, background noise signal-to-noise ratio indicators, text clarity indicators, character matching degree indicators, and speech feature deviation indicators, can transform the physical defects or content recognition uncertainties of different types of evidence, such as images, text, and speech, into calculable values, thus laying the foundation for an objective assessment of evidence quality. It is precisely because qualitative causal types are transformed into refined quantitative data that the assessment of evidence reliability no longer remains superficial. Based on this, the solution further maps these quantitative indicators to corresponding uncertainty impact levels, as the degree of uncertainty impact, according to these quantitative indicators and pre-set judicial evidence reliability assessment standards. This mapping process combines quantitative data with the reliability requirements of judicial practice, allowing the quality of evidence to be directly correlated with its impact on the determination of logical conflicts. For example, a highly distorted image will have a higher quantitative index, thus being mapped to a higher level of uncertainty. This directly suggests that such evidence needs to be given lower weight or undergo more in-depth verification when determining logical conflicts. Through this quantitative assessment, this scheme can precisely identify and differentiate the specific impact of different evidence quality issues on the determination of logical conflicts. This allows for a more accurate elimination or reduction of interference from low-quality evidence when subsequently identifying the direct source of logical conflict, making the final determination of logical conflicts more reliable and persuasive. This quantitative assessment mechanism, closely integrated with the previous steps of obtaining key information, identifying secondary evidence, and determining the source of conflict based on regional attributes, forms a complete chain from macro-level evidence correlation to micro-level quality assessment, effectively improving the accuracy and reliability of judicial evidence data analysis.

[0093] In some preferred embodiments, when the system identifies a logical conflict between the first and second pieces of evidence, and has already obtained the regional attribute causal type of key information in the second evidence—for example, the causal type of a certain key information regional attribute is identified as "image geometric distortion" and "uneven illumination"—the system first obtains the corresponding quantitative indicators to assess the degree of uncertainty in the determination of the logical conflict. For "image geometric distortion," an image processing algorithm based on Hough transform or Fourier transform can be used to calculate the curvature of lines or edges in the image, obtaining a distortion index from 0 to 100, for example, a distortion index of 85. For "uneven illumination," the variance of the image brightness histogram or local contrast analysis method can be used to calculate the brightness difference in different regions of the image, obtaining an illumination deviation index from 0 to 100, for example, an illumination deviation index of 70. After obtaining these quantitative indicators, the system maps these quantitative indicators to the corresponding uncertainty impact level according to a preset judicial evidence reliability assessment standard. The preset judicial evidence reliability assessment standard can be a multi-dimensional scoring model, which includes thresholds and weights for different quantitative indicators. For example, the standard could define a distortion index greater than 80 as "high uncertainty," 60-80 as "medium uncertainty," and less than 60 as "low uncertainty"; similarly, an illumination deviation index greater than 65 would be "high uncertainty," 40-65 as "medium uncertainty," and less than 40 as "low uncertainty." Based on this, a distortion index of 85 would be mapped to a "high uncertainty impact level," and an illumination deviation index of 70 would be mapped to a "high uncertainty impact level." Ultimately, the system can synthesize these levels, for example, by taking the highest level or a weighted average, to determine that the final uncertainty impact of this key information on the logical conflict determination is "high." In this way, the system can transform abstract issues of evidence quality into concrete, actionable evaluation results, providing data support for subsequent conflict source determination.

[0094] This application further proposes steps for mapping quantitative indicators and pre-set judicial evidence reliability assessment standards to corresponding uncertainty impact levels, including:

[0095] Obtain current case attribute information; current case attribute information includes case type, nature of evidence, and focus of court examination.

[0096] Based on the case type, the nature of the evidence, and the focus of the court's examination, the weight parameters of the preset judicial evidence reliability assessment standard are adjusted to obtain the adjusted judicial evidence reliability assessment standard.

[0097] Based on the quantitative indicators and the adjusted judicial evidence reliability assessment standards, the quantitative indicators are mapped to the corresponding uncertainty impact levels.

[0098] Among them, the current case attribute information refers to the specific background and focus information of the case during the judicial trial process. Specifically, this may include the legal classification of the case, such as criminal, civil, or administrative cases; the physical or logical form of the evidence, such as video, audio, documents, or electronic data; and the aspects that the court emphasizes regarding the authenticity, relevance, or legality of the evidence during the examination phase. Its purpose is to provide a personalized basis for the subsequent dynamic adjustment of evidence evaluation standards. The preset judicial evidence reliability evaluation standards refer to a set of general evaluation criteria established within the system before conducting evidence reliability evaluations. These standards may include initial evaluation thresholds or scoring rules for different quantitative indicators, such as image geometric distortion indicators and text clarity indicators. Their purpose is... The purpose is to provide a benchmark for assessing the reliability of evidence; the weighting parameter refers to the numerical factor used to measure the importance of different quantitative indicators or assessment dimensions in the preset judicial evidence reliability assessment standard. It can be expressed as a floating-point number or a percentage. Its purpose is to change the influence of each indicator in the final assessment result by adjusting these parameters in order to adapt to the assessment needs of different cases; the uncertainty impact level refers to the graded judgment of the reliability of evidence based on the quantitative indicators of the evidence and the adjusted assessment standard. It can be divided into multiple discrete levels, such as "low impact", "medium impact", "high impact" or specific numerical ranges. Its purpose is to intuitively represent the potential impact of defects in the evidence on the determination of logical conflicts.

[0099] In some preferred embodiments, suppose that when handling a civil compensation case involving a traffic accident, it is necessary to assess the reliability of a dashcam video as evidence. First, the system obtains the current case attribute information. In this example, the case type is "civil compensation," the evidence type is "video file," and the court's focus during cross-examination may be on the "authenticity of the event" and "liability determination." Based on this case attribute information, the system dynamically adjusts the weight parameters of the preset judicial evidence reliability assessment standard. For example, since it is a traffic accident case and emphasizes the authenticity of the event, the system may increase the weight of quantitative indicators directly related to video quality, such as image geometric distortion, illumination deviation, and background noise signal-to-noise ratio, to ensure that the visual credibility of the video content is fully considered. Simultaneously, for any possible audio, such as in-car conversations, or text information, such as road sign text, the corresponding text clarity or speech feature deviation indicators may be relatively reduced, as their importance in such cases may not be as great as the video footage itself. After adjusting the weight parameters, the system obtains the adjusted judicial evidence reliability assessment standard. Subsequently, various quantitative indicators of the dashcam video evidence, such as the degree of geometric distortion, illumination uniformity, and background noise level of video frames calculated by image processing algorithms, are input into this adjusted evaluation standard. The system then maps these quantitative indicators to corresponding uncertainty impact levels based on the adjusted evaluation standard. For example, if the video exhibits severe illumination unevenness or geometric distortion at a critical moment, even if its original quantitative indicator values ​​are not high, the increased weighting of the illumination deviation and image geometric distortion indicators may result in a final uncertainty impact level classified as "moderate impact" or "high impact," thus accurately reflecting the reliability risk of the video evidence in the current case context. In this way, the evaluation results can more accurately guide the court in accepting and examining evidence.

[0100] This application further proposes steps for adjusting the weighting parameters in the pre-defined judicial evidence reliability assessment criteria, including:

[0101] Based on the current case attribute information, match parameter adjustment rules in the preset rule set, and modify the weight parameters based on the parameter adjustment rules; the preset rule set includes the correspondence between case type, evidence nature, focus of cross-examination and parameter adjustment values.

[0102] The preset rule set refers to a data structure that stores the correspondence between case type, evidence nature, focus of cross-examination, and parameter adjustment values. Specifically, it can be a database, a lookup table, or a set of conditional judgment logic. Its purpose is to provide customized weight adjustment basis for different case attributes. Matching parameter adjustment rules refers to the process of searching and identifying the most suitable or related parameter adjustment rules in the preset rule set based on the specific attributes of the current case. Specifically, it can be done through exact matching, fuzzy matching, or similarity-based calculation. Its purpose is to obtain a weight adjustment scheme applicable to the current case. Parameter adjustment rules refer to specific instructions or numerical combinations obtained from the preset rule set to guide the modification of weight parameters. Specifically, it can be a direct adjustment of values, adjustment ratios, or adjustment functions. Its purpose is to achieve refined adjustment of the weight parameters of the evaluation criteria.

[0103] This application's solution introduces a dynamic adjustment mechanism, enabling the judicial evidence reliability assessment standards to adaptively optimize based on the specific circumstances of each case. Specifically, when assessing evidence reliability, the system first acquires the current case's attribute information, encompassing the case type, the nature of the evidence, and the focus of courtroom examination. Subsequently, the system uses this case attribute information to intelligently match it against a pre-built set of rules. This set of rules predetermines the correlation between different combinations of case attributes and corresponding parameter adjustment values. Through this matching process, the system can identify the parameter adjustment rules most suitable for the characteristics of the current case. Once a suitable rule is matched, the system modifies the weight parameters in the pre-defined judicial evidence reliability assessment standards accordingly. This rule-matching-based weight parameter adjustment method ensures that the assessment standards are no longer fixed but can flexibly adjust the weights of various assessment indicators based on different case types (criminal, civil, etc.), different evidence types (physical evidence, oral evidence, etc.), and different examination focuses (authenticity, relevance, legality, etc.). For example, in a criminal case focusing on the authenticity of evidence, the system can automatically increase the weight of physical defect characteristics such as image geometric distortion and text clarity; while in a civil case focusing on relevance, it may increase the weight of content feature analysis results (such as recognition uncertainty caused by voice accent). This refined weight adjustment, combined with the previous step of mapping quantitative indicators to corresponding uncertainty impact levels based on quantitative indicators and the adjusted judicial evidence reliability assessment standards, forms a more complete assessment process. It overcomes the limitations of using uniform assessment standards or simple preset parameter sets, making the mapping from quantitative indicators to uncertainty impact levels more closely aligned with the needs of actual cases, thereby significantly improving the accuracy and applicability of evidence reliability assessment. It is precisely because of this dynamic and intelligent weight adjustment mechanism that the assessment results can more accurately reflect the authenticity and reliability of evidence in specific judicial contexts, providing a more valuable basis for courtroom examination.

[0104] In some preferred embodiments, it is assumed that there exists a preset set of rules, stored in the form of structured data, such as a JSON file or a relational database table. This set contains multiple rules, each defining a specific combination of case type, evidence nature, and examination focus, and corresponding to a set of parameter adjustment values. For example, a rule might be defined as follows: when the case type is "criminal case," the evidence nature is "video evidence," and the examination focus is "authenticity," the corresponding parameter adjustment values ​​are "increase the weight of the image geometric distortion index by 0.2, and increase the weight of the illumination deviation index by 0.1." When it is necessary to assess the reliability of a piece of evidence, the system first obtains the attribute information of the current case. For example, the current case is identified as a "criminal case," the evidence involved is "video evidence," and the focus of court examination is "authenticity." The system uses this attribute information to match within the preset set of rules. This matching process can be a database query operation or iterating through the rule entries in the JSON file to find parameter adjustment rules that completely match or are highly similar to the attribute information of the current case. Once one or more parameter adjustment rules are matched, the system modifies the weight parameters in the preset judicial evidence reliability assessment criteria based on these rules. For example, if the rule "criminal case - video evidence - authenticity" is matched, the system will increase the weight of the "image geometric distortion degree index" by 0.2 and the weight of the "illuminance deviation index" by 0.1. These adjusted weight parameters are then applied to the mapping process of the quantitative indicators, thereby obtaining an uncertainty impact level that better reflects the actual situation of the current case. In this way, the assessment criteria can be customized according to the subtle differences in the case, making the assessment results more targeted and accurate.

[0105] In some embodiments described above in this application, a method is proposed to match parameter adjustment rules in a preset rule set based on current case attribute information, and then modify weight parameters based on these rules. Specifically, this method can involve identifying a single key attribute of the case, such as case type, then searching for a parameter adjustment rule directly corresponding to that case type in the preset rule set, and modifying the weight parameters according to that single rule. This allows for preliminary adjustments to the evaluation criteria based on the initial case classification. However, in practice, case attribute information can be complex, and simply matching a single rule may not fully utilize the information in the preset rule set, resulting in inaccurate parameter adjustments and affecting the accuracy of the judicial evidence reliability assessment.

[0106] In response, this application further proposes the following steps: adjusting matching parameters within a pre-defined rule set based on current case attribute information, and modifying weight parameters based on these adjustment rules.

[0107] Based on the case type, evidence nature, and focus of cross-examination corresponding to the current case attribute information, multiple first rules matching the current case attribute information are queried from the preset rule set;

[0108] Determine the degree of matching between each first rule and the current case attribute information, and weight the parameter adjustment values ​​corresponding to multiple first rules according to the degree of matching to determine the final parameter adjustment value, which is used to adjust the weight parameters in the preset judicial evidence reliability assessment standard.

[0109] Among them, multiple first rules refer to rules in the preset rule set that are related to or meet some conditions of the current case attribute information (including case type, nature of evidence, and focus of cross-examination). These rules can be identified according to the preset matching logic, and their purpose is to provide multi-dimensional reference information for subsequent parameter adjustments. The matching degree refers to the correlation strength or conformity between each first rule and the current case attribute information. It can be determined by calculating the similarity between the rule and the case attributes, the number of attributes that conform, or the preset priority. Its purpose is to quantify the applicability of each rule to the current case. The parameter adjustment value refers to the specific value or proportion preset by each first rule for modifying the weight parameters in the judicial evidence reliability assessment standard. It can be a fixed value, a range value, or a calculation formula. Its purpose is to provide a basis for the fine-tuning of the weight parameters.

[0110] This application's solution addresses the problem of inaccurate parameter adjustments caused by single-rule matching when case attribute information is complex by introducing a multi-rule matching and weighted adjustment mechanism. Specifically, when adjusting the weight parameters in the preset judicial evidence reliability assessment criteria, the system no longer simply searches for a single matching rule. Instead, based on the current case attribute information, such as case type, evidence nature, and the focus of courtroom examination, it queries the preset rule set for multiple first rules that match the current case attribute information. This process ensures a comprehensive consideration of case attributes and avoids biases caused by information omissions. Subsequently, the system determines the degree of matching between each first rule and the current case attribute information. For example, one rule may highly match the case type, while another rule may be more relevant to the focus of examination; their respective degree of matching will differ. Based on these degree of matching, the system weights the parameter adjustment values ​​corresponding to multiple first rules. This means that the higher the degree of matching, the greater the proportion of the parameter adjustment value in the final adjustment result. Through this weighting method, the influence of multiple rules can be integrated to obtain a more accurate and detailed final parameter adjustment value. This final parameter adjustment value is then used to adjust the weighting parameters in the pre-set judicial evidence reliability assessment standards. This refined adjustment of weighting parameters makes the subsequent process of mapping quantitative indicators to uncertainty impact levels more accurate, thereby improving the reliability of assessing the degree of uncertainty impact of various key pieces of information on the determination of logical conflicts. Ultimately, it enables a more accurate identification of the source of direct logical conflicts, enhancing the overall accuracy and reasonableness of judicial evidence reliability assessment. This multi-rule weighted adjustment mechanism allows the judicial evidence reliability assessment standards to adapt more flexibly and accurately to the complexity and specificity of different cases, thus providing more valuable evidence reliability judgments in judicial practice.

[0111] In some preferred embodiments, it is assumed that a judicial evidence analysis system needs to dynamically adjust the weight parameters in its internal judicial evidence reliability assessment criteria based on the attribute information of the current case. When the system receives the attribute information of a case, for example, the case is identified as a "criminal case," the nature of the evidence is "confession evidence," and the focus of court examination is "authenticity," the system will first query a preset set of rules. This preset set of rules can be a database or a configuration table, which stores a large number of predefined rules, each rule being associated with a specific combination of case attributes and corresponding parameter adjustment values.

[0112] The system will search the rule set for all matching rules based on the three attributes of the current case: "criminal case," "confession evidence," and "authenticity." For example, the system may find multiple first rules: one rule might apply to the combination of "criminal case" and "confession evidence," another rule might apply to the combination of "confession evidence" and "authenticity," and yet another rule might apply to the combination of "criminal case" and "authenticity." These are all multiple first rules that match the attribute information of the current case.

[0113] Next, the system determines the degree of match between each first rule and the current case's attribute information. This can be calculated based on the number of matched attributes, the importance of the attributes, or a preset matching algorithm. For example, if a first rule completely matches all three attributes of the case, its degree of match can be set to the highest; if it only matches two attributes, the degree of match will be reduced accordingly.

[0114] The system then weights the parameter adjustment values ​​corresponding to each of the first rules based on these matching degrees. For example, if the matching degree of the first rule is 0.9, its parameter adjustment value is +0.1; the matching degree of the second rule is 0.8, its parameter adjustment value is +0.05; and the matching degree of the third rule is 0.7, its parameter adjustment value is -0.02. The final parameter adjustment value can then be calculated as the weighted average of (0.9*0.1) + (0.8*0.05) + (0.7*-0.02). This calculated final parameter adjustment value is subsequently used to precisely adjust the weight parameters in the preset judicial evidence reliability assessment criteria, thereby enabling the assessment criteria to more accurately reflect the specific needs and complexities of the current case.

[0115] Furthermore, this application proposes a judicial evidence storage data analysis system, wherein the judicial evidence storage data includes video files, audio files, and / or document files; such as Figure 2 As shown, the system includes:

[0116] The mapping establishment module 201 is used to determine the key information of the inherent logical relationship of the event based on the judicial evidence data corresponding to the current case, and to establish a mapping relationship between the key information and the judicial evidence data and physical storage address; the key information includes the time, place, people and / or behavior information of the event.

[0117] The determination module 202 is used to determine the second evidence based on the characteristic information, mapping relationship and preset rules of the first evidence currently specified by the user; the second evidence is other evidence in the judicial evidence storage data that has a direct logical conflict with the first evidence;

[0118] The generation module 203 is used to generate and display logical contradiction information between the first piece of evidence and the second piece of evidence. The logical contradiction information is a contradiction chain diagram with the first piece of evidence node as the central node and the second piece of evidence node radiating outwards. The first piece of evidence node and the second piece of evidence node in the contradiction chain diagram are both linked to the original evidence content through mapping relationships.

[0119] Through the above technical solution, this application provides a judicial evidence storage data analysis system that transforms abstract judicial evidence storage data analysis methods into practically operable tools. Through modular design, the system achieves automated processing of massive amounts of judicial evidence storage data, extraction and mapping of key information, and intelligent identification and visualization of logical conflicts between pieces of evidence. This effectively solves the problem that relying solely on analytical methods cannot complete judicial evidence storage data analysis in practical applications, providing judicial personnel with a platform to improve the efficiency and accuracy of evidence analysis. This enables them to quickly locate and understand contradictions between pieces of evidence, assist in judicial decision-making, and improve the efficiency and fairness of judicial trials.

[0120] The above description is merely an embodiment of this application and is not intended to limit the scope of protection of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method for analyzing judicial evidence storage data, characterized in that, Judicial evidence storage data includes video files, audio files, and / or document files; the method includes: Based on the judicial evidence data corresponding to the current case, key information about the inherent logical connection of the event is determined, and a mapping relationship is established between the key information, the judicial evidence data, and the physical storage address; the key information includes the time, place, people, and / or behavior information of the event. Based on the characteristic information of the first piece of evidence currently specified by the user, the mapping relationship, and the preset rules, the second piece of evidence is determined; the second piece of evidence is other evidence in the judicial evidence storage data that has a direct logical conflict with the first piece of evidence. Generate and display logical contradiction information between the first piece of evidence and the second piece of evidence; the logical contradiction information is a contradiction chain diagram with the first piece of evidence node as the central node and the second piece of evidence node radiating outwards; the first piece of evidence node and the second piece of evidence node in the contradiction chain diagram are all linked to the original evidence content through the mapping relationship; The step of determining the second evidence based on the characteristic information of the first evidence currently specified by the user, the mapping relationship, and preset rules includes: Obtain the regional attributes of the key information corresponding to the second piece of evidence; Based on the aforementioned regional attributes, determine the source of the direct logical conflict; The regional attributes for obtaining the key information corresponding to the second evidence include: Physical defect feature analysis is performed on the area where the key information corresponding to the second evidence is located to identify image geometric distortion, uneven lighting and / or background noise caused by the evidence carrier or the collection environment, and the physical defect feature analysis results are obtained. Content feature analysis is performed on the area where the key information corresponding to the second piece of evidence is located to identify recognition uncertainties caused by text blurring, font abnormalities and / or speech accents, and the content feature analysis results are obtained. Based on the analysis results of physical defect characteristics and content characteristics, the causal type of the regional attribute is determined and used as the regional attribute.

2. The judicial evidence storage data analysis method according to claim 1, characterized in that, The characteristic information of the first piece of evidence includes the time, place, people and / or behavior information of the event described in the first piece of evidence; the preset rule is a legal logic contradiction rule; The step of determining the second evidence based on the characteristic information of the first evidence currently specified by the user, the mapping relationship, and preset rules includes: Based on the characteristic information of the first evidence and the preset rules, other evidence that has a direct logical conflict with the first evidence is found in the key information of the mapping relationship and identified as the second evidence.

3. The judicial evidence storage data analysis method according to claim 1, characterized in that, Based on the aforementioned regional attributes, the source of the direct logical conflict is determined, including: Obtain the cause type of the regional attributes corresponding to each key piece of information involved in the logical conflict; Based on the cause type of the regional attributes corresponding to each key piece of information, assess the degree of uncertainty in the determination of logical conflicts of each key piece of information. The source of the direct logical conflict is determined based on the degree of impact of the uncertainty.

4. The judicial evidence storage data analysis method according to claim 3, characterized in that, Based on the causal type of the regional attributes corresponding to each key piece of information, assess the degree of uncertainty in the determination of logical conflicts caused by each key piece of information, including: Based on the key information involved in the logical conflict, obtain the quantitative indicators of the regional attribute cause type corresponding to the key information; the quantitative indicators include image geometric distortion degree indicator, illumination brightness deviation indicator, background noise signal-to-noise ratio indicator, text clarity indicator, character matching degree indicator and / or speech feature deviation indicator. Based on the quantitative indicators and the preset judicial evidence reliability assessment standards, the quantitative indicators are mapped to the corresponding uncertainty impact levels, which are used as the degree of uncertainty impact.

5. The judicial evidence storage data analysis method according to claim 4, characterized in that, Based on the quantitative indicators and the preset judicial evidence reliability assessment standards, the quantitative indicators are mapped to corresponding uncertainty impact levels, which are used as the degree of uncertainty impact, including: Obtain current case attribute information; the current case attribute information includes case type, nature of evidence, and focus of court examination. Based on the case type, the nature of the evidence, and the focus of the court's examination, the weight parameters in the preset judicial evidence reliability assessment standard are adjusted to obtain the adjusted judicial evidence reliability assessment standard. Based on the quantitative indicators and the adjusted judicial evidence reliability assessment standards, the quantitative indicators are mapped to the corresponding uncertainty impact levels.

6. The judicial evidence storage data analysis method according to claim 5, characterized in that, The adjustment of the weight parameters in the preset judicial evidence reliability assessment standard includes: Based on the current case attribute information, matching parameter adjustment rules are performed in a preset rule set, and the weight parameters are modified based on the parameter adjustment rules; the preset rule set includes the correspondence between case type, evidence nature, focus of cross-examination and parameter adjustment values.

7. The judicial evidence storage data analysis method according to claim 6, characterized in that, Based on the current case attribute information, matching parameter adjustment rules are performed in a preset rule set, and the weight parameters are modified based on the parameter adjustment rules, including: Based on the case type, evidence nature, and focus of cross-examination corresponding to the current case attribute information, multiple first rules matching the current case attribute information are queried from the preset rule set; Determine the degree of matching between each of the first rules and the current case attribute information, and weight the parameter adjustment values ​​corresponding to multiple first rules according to the degree of matching to determine the final parameter adjustment value, which is used to adjust the weight parameters in the preset judicial evidence reliability assessment standard.

8. A judicial evidence preservation data analysis system, characterized in that, Judicial evidence storage data includes video files, audio files, and / or document files; the system includes: The mapping establishment module is used to determine key information about the inherent logical relationship of an event based on the judicial evidence data corresponding to the current case, and to establish a mapping relationship between the key information and the judicial evidence data and physical storage address; the key information includes the time, place, people and / or behavior information of the event. The determination module is used to determine the second evidence based on the characteristic information of the first evidence currently specified by the user, the mapping relationship, and preset rules; the second evidence is other evidence in the judicial evidence storage data that has a direct logical conflict with the first evidence; The step of determining the second evidence based on the characteristic information of the first evidence currently specified by the user, the mapping relationship, and preset rules includes: Obtain the regional attributes of the key information corresponding to the second piece of evidence; Based on the aforementioned regional attributes, determine the source of the direct logical conflict; The regional attributes for obtaining the key information corresponding to the second evidence include: Physical defect feature analysis is performed on the area where the key information corresponding to the second evidence is located to identify image geometric distortion, uneven lighting and / or background noise caused by the evidence carrier or the collection environment, and the physical defect feature analysis results are obtained. Content feature analysis is performed on the area where the key information corresponding to the second piece of evidence is located to identify recognition uncertainties caused by text blurring, font abnormalities and / or speech accents, and the content feature analysis results are obtained. Based on the analysis results of physical defect characteristics and content characteristics, the causal type of the regional attribute is determined and used as the regional attribute. The generation module is used to generate and display logical contradiction information between the first piece of evidence and the second piece of evidence; the logical contradiction information is a contradiction chain diagram with the first piece of evidence node as the central node and the second piece of evidence node radiating outwards; the first piece of evidence node and the second piece of evidence node in the contradiction chain diagram are both linked to the original evidence content through the mapping relationship.

Images