16 results about "Evidence analysis" patented technology

The invention discloses a traffic accident video evidence analysis system and method based on multi-modal deep learning, and relates to the technical field of deep learning, and the method comprises the steps: carrying out the target recognition and tracking through radar point cloud, extracting a target number component and a motion variance component, carrying out the calculation through normalization and linear combination, and obtaining a dynamic fusion weight of each modal; and quantization and adaptive adjustment of scene complexity are realized. Then extracting each modal feature vector in a preset time sequence window, and performing weighted fusion based on the weight to form a unified feature vector; and inputting the uniform feature vector into a pre-trained accident classification model, and outputting whether an accident occurs or not and a type result. The method has the advantages that through complementary enhancement of multi-modal data, the accident identification accuracy in complex scenes such as low illumination and rain and fog is improved; adaptive adaptation to different scenes is realized through dynamic weight distribution; the whole scheme forms a closed loop, and has high robustness and practical value.

The invention provides a private encryption type ENC electronic file analysis evidence obtaining method and system, relates to the technical field of computer electronic evidence obtaining, and solves the problem of case-related evidence chain breakage caused by difficulty in evidence obtaining of undisclosed private encryption type files in the prior art. The method comprises the following steps: firstly, initializing a target system in a hardware isolation environment, positioning and extracting a binary static feature sample, performing semantic analysis, dividing a feature function chain boundary, and establishing a binary module association index to extract a decryption module; based on Session structured dynamic taint analysis, a specific pollution source is tracked, and variable propagation, cross check password generation logic and traceability password check execution flow are self-defined; and finally, through dynamic anchor point positioning based on hardware breakpoints, reconstructing and restoring a private encryption and decryption protocol, and outputting a matrix for evaluating and verifying the evidence obtaining feasibility of the ENC electronic file. According to the method, the evidence obtaining and analysis process does not depend on an original software system, and complete autonomy and controllability in technology are achieved.

According to the automobile electronic data forensics analysis method and device and the storage medium, EDR and VDR data are extracted through a portable automobile electronic data forensics device, and on the basis of the following analysis processes, two automobile driving track points with unified time and space are automatically generated in combination with field investigation data, so that quick reproduction analysis of accident tracks at a software side is realized; the analysis process comprises the steps that a Cartesian simulation coordinate system of the passenger car is established, and simulation track points are generated; based on Kruger-n series expansion of a WGS84 ellipsoid, mapping track points of a simulation coordinate system to a geographic coordinate system through a mapping relation between simulation coordinates corresponding to an anchor point, namely a passenger car collision point and corresponding GPS coordinates; performing affine transformation from a Cartesian simulation coordinate system to a UTM coordinate system; and back projection is carried out, i.e., the UTM coordinate system to the latitude and longitude. According to motor vehicle accident data evidence obtaining and accident analysis, the accident investigation evidence obtaining and analysis efficiency is improved through evidence obtaining tool innovation and accident track reconstruction analysis method innovation.

The application relates to the technical field of binary program analysis and rewriting, and discloses an immediate number symbolization judgment method and system of a binary rewriting program, which comprises a preprocessing step, a basic evidence analysis step, an intermediate evidence analysis step, a high-level evidence analysis step and a fusion decision step. The preprocessing layer carries out structural analysis and semantic modeling on a binary file, extracts candidate immediate numbers and meta information, the basic evidence layer carries out preliminary weighting based on explicit rules, the intermediate evidence layer carries out quantitative evaluation through heuristics and statistical patterns, the high-level evidence layer carries out high-precision analysis on the immediate numbers with semantic ambiguity, the fusion decision layer summarizes the evidences at all levels, forms a comprehensive tendency value through weighted calculation, classifies the immediate numbers, and generates a final symbolization strategy. The application significantly improves the accuracy and interpretability of symbolization judgment, balances the analysis precision and processing efficiency, and is suitable for binary files of different compilers, platforms and file stripping states.

The system according to the embodiment aims to efficiently and accurately perform risk analysis and verdict prediction in civil trials. [Solution] A system according to an embodiment includes a risk analysis unit, an evidence analysis unit, a verdict analysis unit, and a simulation provision unit. The risk analysis unit identifies risks. The evidence analysis unit evaluates evidence based on the risks identified by the risk analysis unit. The verdict analysis unit predicts a verdict based on the evidence evaluated by the evidence analysis unit. The simulation provision unit provides a simulation based on the verdict outcome predicted by the verdict analysis unit.

The invention discloses an off-site auditing method and system based on artificial intelligence, and belongs to the technical field of off-site auditing intelligence, and the method comprises the steps of data engineering, auditing evidence analysis, customer risk assessment and comprehensive report generation. According to the method, auditing evidence analysis is carried out by adopting a cross-modal evidence chain intelligent analysis method, unified modeling is carried out on business records of different sources and different forms, the potential relation between evidences is mined, an evidence chain capable of truly reflecting the business process is formed, and the business efficiency is improved. Therefore, risk features and abnormal modes hidden behind various data are revealed more comprehensively. A customer risk assessment method based on potential risk reasoning is adopted to carry out customer risk assessment, scattered abnormal evidences are aggregated to a customer level, overall customer risk features are generated, a potential risk conduction relation between customers is deduced, high-risk customers and potential related parties thereof are dynamically identified, more accurate and comprehensive customer risk assessment is realized, and the customer risk assessment efficiency is improved. And thus, the off-site auditing efficiency, comprehensiveness and accuracy are improved.

The invention relates to an audit data processing method and device and computer equipment. The method comprises the steps of obtaining an auditing template matched with an auditing risk signal from an auditing knowledge graph under the condition that the auditing risk signal is detected, and determining a corresponding evidence obtaining task list according to the auditing template; for each evidence obtaining task in the evidence obtaining task list, calling an audit evidence text matched with the evidence obtaining task from the service system; retrieving auditing knowledge matched with the auditing evidence text from an auditing knowledge graph, and constructing context information according to the auditing knowledge and the auditing evidence text; inputting the context information into an evidence analysis model to obtain a corresponding evidence analysis result; the evidence analysis result comprises at least one of the abstract of the context information, the key information and the compliance judgment. By adopting the method, the auditing efficiency can be improved.

The invention belongs to the technical field of video analysis, and provides a network live broadcast evidence obtaining analysis method, which comprises the following steps: S1, monitoring a specified target in real time, and obtaining a broadcast starting state and a live broadcast stream; s2, extracting the audio of a specified target; s3, performing voice recognition on the audio extracted in the step S2 to obtain a live broadcast text draft; and S4, analyzing the semantic characters obtained in the step S3. The invention further correspondingly provides a network live broadcast evidence obtaining and analyzing system. According to the invention, intelligent evidence obtaining and analysis integration can be realized, automatic office is realized, manpower is reduced, and working efficiency is improved.

The invention provides an electronic data semantic abstracting and evidence sorting method combined with a large language model, and relates to the technical field of artificial intelligence and electronic data analysis, and the method comprises the steps: S1, data collection: collecting text data from a mobile phone, a computer and social platform equipment, and obtaining source data; through semantic understanding and dynamic antagonism analysis of a large language model, deep mining of evidence values is realized, specific meanings of anaphora relationships can be accurately analyzed, key information omission is avoided, potential intentions and emotional tendencies implied in texts are analyzed, real purposes under surface characters are revealed, and the method and the system have a good application prospect. According to the method, contradictory points of different evidences in timelines, fact declaration and logic can be automatically identified, misjudgment caused by inconsistency in an evidence chain is effectively avoided, analysis work is converted into active and deep logical reasoning and verification from passive and shallow information retrieval, and the insight and accuracy of evidence analysis are greatly improved.

The invention relates to the technical field of binary program analysis and rewriting, and discloses an immediate operand symbolization judgment method and system for a binary rewriting program, and the method comprises a preprocessing step, a basic evidence analysis step, an intermediate evidence analysis step, an advanced evidence analysis step and a fusion decision-making step. The preprocessing layer is used for carrying out structure analysis and semantic modeling on the binary file and extracting candidate immediate operands and meta-information; performing preliminary weighting on the basic evidence layer based on an explicit rule; the intermediate evidence layer carries out quantitative evaluation through a heuristic mode and a statistical mode; the advanced evidence layer performs high-precision analysis on the immediate data of semantic ambiguity; and the fusion decision-making layer summarizes the evidences of all levels, forms a comprehensive tendency value through weighted calculation, classifies immediate operands, and generates a final symbolization strategy. According to the method, the accuracy and interpretability of symbolization judgment are remarkably improved, the analysis precision and the processing efficiency are both considered, and the method is suitable for binary files in different compilers, platforms and file stripping states.

The application discloses a traffic accident video evidence analysis system and method based on multi-modal deep learning, relates to the technical field of deep learning, and first utilizes radar point clouds to perform target identification and tracking, extracts target quantity components and motion variance components, and calculates dynamic fusion weights of each mode through normalization and linear combination, so that the quantification and adaptive adjustment of scene complexity are realized. Then, each mode feature vector is extracted in a preset timing window, and weighted fusion is performed based on the weights to form a unified feature vector. The unified feature vector is input into a pre-trained accident classification model, and an accident occurrence and type result are output. The method has the advantages that: through complementary enhancement of multi-modal data, the accident identification accuracy in complex scenes such as low light, rain and fog is improved; adaptive adaptation to different scenes is realized through dynamic weight distribution; the overall scheme forms a closed loop, and has high robustness and practical value.

The application relates to the technical field of administration, and discloses a network public opinion hot event intelligent identification and quick handling system based on a cross-large model, which comprises the following: in view of the problem that public perception positions are deviated from real source positions caused by hidden leakage of a gas pipe network in a real entity space, further leading to multi-source public opinion misjudgment and work order misplacement, the scheme accesses public opinion data flow, utilizes a double-large model to perform heterogeneous evidence analysis and multi-dimensional quantification, and outputs expected geographic coordinates and perception dominance; combined with pipe network topology and double-medium propagation time modeling, the scheme performs offset evaluation on a public sniffer guide misplaced ring belt; based on evaluation values and evolution residuals, the scheme performs particle swarm optimization to inversely calculate real leakage source positions; finally, the scheme takes the inversely calculated source positions as anchor points to perform hot spot reconstruction and fission inhibition, and calculates comprehensive emergency degrees to generate a quick handling work order sequence. The application realizes accurate alignment of public opinion evolution and real disaster evolution, and greatly improves the accuracy and response efficiency of emergency quick handling.

One or more embodiments of the specification provide a judgment generation method and system, which realizes the alignment of artificial intelligence reasoning path and legal argumentation logic by decomposing the trial thinking of a judge into a chain-verification trial process of "fact structuring-dispute focus identification-evidence analysis-law application-judgment generation". The input and output of each stage are objectified and structured, so that the entire judgment generation process is "visible", the decision basis is clear, the reasoning steps are traceable, and a controllable, reliable, rigorous, universal and transparent intelligent auxiliary trial mechanism is built.

This application provides a method and apparatus for superedge evidence analysis of multi-site chromatin interactions, relating to the field of high-order graph data processing. The method includes: decomposing and mapping target reads in multi-connected read interaction data to a set of anchor points; normalizing the hit anchor points to generate superedge signatures and saving evidence pointers to form superedge evidence transactions; incrementally mining the superedge evidence transactions under biological and statistical constraints to obtain a set of candidate superedges; backtracking read evidence based on evidence pointers and calculating the consistency statistic of the target candidate superedges; constructing a significance discrimination null model based on the consistency statistic, outputting a significance probability value, and obtaining a corrected significance probability value through multiple tests; and outputting the superedge evidence analysis results and a verifiable data package for backtracking evidence pointers based on the corrected significance probability value. This application solves the problem of inaccurate significance determination results in existing analysis systems when evaluating multi-site interactions.

One or more embodiments of the invention provide a discriminant generation method and system, and the method and system achieve the alignment of an artificial intelligence reasoning path and a legal argument logic through decomposing the judgment thinking of a judge into a chained verifiable judgment process of fact structuring, dispute focus recognition, evidence analysis, law application and judgment generation. The input and output of each stage are concrete and structured, so that the whole judgment generation process is visible, the decision basis is clear, the reasoning steps are traceable, and a controllable, reliable, rigorous, universal and transparent intelligent auxiliary judgment mechanism is constructed.

The invention provides an abnormal transaction behavior analysis, research and judgment method and system based on multi-source data, and relates to the technical field of abnormal transaction data analysis, and the method comprises the steps: constructing a multi-source account association topology, and generating a collection instruction set; obtaining multi-source heterogeneous fund transaction data and unifying the formats; extracting features to construct a data quality evaluation vector, and calculating a credibility weight for hierarchical labeling; constructing a weighted fund flow network structure; abnormal transaction paths are identified and scored, and the highest confidence coefficient abnormal transaction sub-graph is screened out; and generating an evidence analysis result. According to the invention, intelligent collection, quality evaluation and credible association analysis of multi-source data are realized, and the collection efficiency and accuracy of transaction crime evidence are improved.