20 results about "Case identification" patented technology

The invention provides a difficult case recognition method, system and device which are used for mining possible difficult case scenes in the vehicle driving process and achieving accurate mining of the difficult case scenes. The method comprises the steps that firstly, a sensing data set collected in the vehicle driving process is obtained, the sensing data set is obtained by collecting a current scene through at least one sensor in a vehicle, and sensing data comprises multiple frames of sensing data; then the multi-frame sensing data is input into a sensing model to obtain a plurality of sensing results, and the sensing model is used for detecting elements in the input data and can output information of the detected elements, such as types, positions, shapes or sizes of the elements; then, according to the similarity among the multiple perception results, a difficulty coefficient is determined, and the difficulty coefficient is used for measuring the difficulty of determining a vehicle driving decision in the current scene or used for measuring the perception ability of the vehicle or the perception model in the current scene.

The application discloses a social network-based insurance fraud gang identification method, system, device and medium, and relates to the technical field of computers, wherein the method comprises the following steps: constructing a social network relationship graph based on the role information and subject information in a newly added claim event and using insurance business relationships; obtaining traversal starting points for constructing a closed-loop relationship graph according to the social network relationship graph; constructing a plurality of closed-loop relationship graphs for a plurality of traversal starting points by using a first adjacent point acquisition method based on claim event relationships; and obtaining a target closed-loop relationship graph by merging the plurality of closed-loop relationship graphs by using a second adjacent point acquisition method based on claim event relationships, wherein the node information in the target closed-loop relationship graph is used to represent a target insurance fraud gang. The application is suitable for efficient investigation of fraud case identification and improves the accuracy and efficiency of fraud case identification.

The invention relates to the technical field of computer information processing, in particular to a trademark three-level text dividing method based on Redis cache, which comprises the following steps of: receiving and normalizing trademark characters, commodity and service information, trademark classification numbers, applicant identifiers and trademark patterns, and writing normalized data and trademark case identifiers into Redis cache; generating trademark case portrait vectors through the multi-modal feature extraction network and the graph neural network; and constructing an input vector in combination with the to-be-examined quantity and the processing duration of the examination queues at all levels, inputting the text division strategy model to obtain a target examination level, and updating the corresponding queues in the cache to realize automatic shunting of trademark cases. The case complexity and the review load are considered in the text division process, the manual case division pressure is relieved, and the risk case recognition capability and the overall review efficiency are improved.

The application discloses a case handling period risk and record compliance risk linkage summary method, system and computer program product under an internal network environment. Unified case identification is used to associate case basic data, period calculation results, record documents, compliance verification reports and operation logs. A period calculation module determines the expiration date, remaining days and period early warning risks from the case filing date, handling period, extension, non-counted period days and handling state. A record compliance verification module uses a hit chain traceable deterministic rule base to obtain record compliance risks from the associated records, and does not rely on the uncertainty output of a large language model. A risk synthesis module maps the above two types of risks with different sources, types and calculation methods into high, medium and low levels, respectively, and takes the higher one to obtain a comprehensive risk level. In response to the events of handling, resuming handling, soft deleting, adding a new record or generating a report, the business semantics are automatically recalculated and refreshed. The whole process is completed on the internal network and the data is not externally transmitted.

The application provides an inpatient hospital infection case identification method and system based on an AI large model, comprising: constructing a screening standard knowledge base based on infection sites, clinical diagnosis, etiological diagnosis and supplementary notes; collecting desensitization diagnosis and treatment data of inpatients, splicing with prompt words to form input text; calling a large model interface to process the input text of the inpatients, and the large model performs reasoning according to the screening standard knowledge base; analyzing the structured data returned by the large model to obtain infection early warning information; wherein the infection early warning information comprises: whether to be infected, infection site, risk level, key evidence and reference standard. The application can provide an inpatient hospital infection case identification method capable of improving the accuracy and scientificity of hospital infection screening, analyzing complex medical screening logic and providing key evidence with clinical interpretability.

A functional case identification method, a functional case, and an electronic device, relating to the technical field of terminals. The method comprises: the electronic device detects a magnetic pole of a functional case worn by the electronic device, and then determines the type of the functional case on the basis of the magnetic pole of the functional case and a mapping relationship, wherein the mapping relationship is a preset mapping relationship between the magnetic pole of the functional case and the type of the functional case. In this way, the electronic device can determine the type of the functional case by identifying the magnetic pole of the worn functional case, thereby not only identifying the type of the functional case, but also improving identification efficiency, thus reducing system power consumption of the electronic device.

The invention discloses an adaptive retrieval enhanced teaching case generation method. Firstly, a knowledge graph strictly aligned with a subject system is constructed, unification is completed in a subject category and case knowledge unit level, and six types of elements including teaching objectives, ideological and political focuses, three forms, exercises, professional items and incentive wakeup are further standardized. Secondly, analyzing teacher requirements by the case identification model, carrying out multi-path retrieval, evaluating the correlation and calculation overhead of candidate paths, and determining an optimal reference path; and then, rewriting the triple in the candidate path into a natural language, generating a corresponding case fragment according to an element driving generation model, implementing quality evaluation based on integrity, clarity, consistency and compatibility, and finally synthesizing a teaching case with a complete structure. The method effectively inhibits model illusion and redundancy generation, significantly improves content accuracy and coherence, and is suitable for application scenarios such as course preparation and intelligent teaching assistance.

The invention provides an automatic test code generation method and device and electronic equipment, and relates to the technical field of automatic test code generation, and the method comprises the following steps: initializing a structured test knowledge base; extracting test case information from the test data source so as to construct a standardized problem dictionary comprising case identifiers, test levels, step types and associated scenes through the test case information; generating a new test keyword code by sequentially executing script-based concurrent preprocessing, recall process concurrent retrieval and in-script concurrent generation on the question dictionary; performing intelligent unified verification on the new test keyword code; calling a preset use case template, filling the question dictionary information and the keyword codes conforming to the target test specification to generate complete test codes, and updating newly generated keywords to the structured test knowledge base; the test code can be automatically and accurately generated.

This invention discloses an intelligent judicial multi-task prediction method based on dynamic difficult case perception. The method constructs a multi-task prediction model, including a shared encoder, a crime and legal provision prediction layer, a difficult case identification module, and an adaptive threshold adjustment mechanism. During the training phase, the model comprehensively evaluates sample difficulty by calculating prediction entropy, normalized prediction dispersion, and cross-task consistency scores, and identifies difficult cases using dynamically adjusted thresholds, assigning them higher training weights. During the inference phase, the model directly outputs crime prediction and legal provision recommendation results. This invention can adaptively identify key difficult cases, effectively alleviate the class imbalance problem in judicial data, and significantly improve the model's prediction performance for tail categories and its overall robustness.

The invention discloses a tobacco inspection capability optimization method, apparatus and device, and a computer program product. The method comprises the steps of identifying a regional concentration region of a false cigarette sample source according to sample source regional information of a false cigarette case, and performing statistics on the total amount of searched cases; determining a sample concentration degree of each region and a training region coverage range, and identifying a training uncovered region; integrating false cigarette case records and personnel training records to obtain a multi-region case library; according to the multi-region case library, false making identification key points of each region are determined; generating a cross-region communication learning plan according to the multi-region false cigarette identification key point list and the region concentrated region, and arranging inspectors to participate in multi-region case training according to the cross-region communication learning plan; and determining the coverage improvement degree according to the inspection coverage record and the false cigarette case identification accuracy, and determining the tobacco inspection capability improvement amplitude of the inspector. The cross-region identification capability of tobacco inspectors is improved, and the method can be applied to the technical field of enterprise management.

The invention discloses a project demand analysis method, device and equipment and a storage medium, and relates to the technical field of project management, and the project demand analysis method comprises the steps that training multi-source project data is acquired, and the training multi-source project data comprises product data, historical index data and system feature data; analyzing the unstructured data of the training multi-source project data to obtain a function implementation process, and converting the function implementation process into a function feature vector; performing normalization processing on the structured data of the training multi-source project data to obtain standard numerical data; and performing function identification training, case identification training and risk weighting training according to the function feature vector and / or the standard numerical data to obtain a demand analysis model, and performing project demand analysis according to the demand analysis model.

The invention relates to a repeated urban management case recognition method based on multi-modal data, and belongs to the field of pattern recognition, and the method comprises the steps: constructing a high-frequency repeated case library; screening a case matched with the type of the to-be-identified case from a high-frequency repeated case library as a to-be-compared case; respectively extracting feature information of the to-be-identified case and the to-be-compared case by adopting a feature extraction network, wherein the feature information comprises crime image features, text description features and geographic position features; fusing the three single-mode features to obtain a comprehensive feature; calculating a similarity result of the two cases according to the comprehensive characteristics of the to-be-identified case and the comprehensive characteristics of the to-be-compared case; and inputting a similarity result into a binary classification network, and determining whether the to-be-identified case is a repeated case or not. According to the method and the system, the accuracy and the processing efficiency of repeated case recognition are remarkably improved, the adaptability of the system to complex scenes is enhanced, and the method and the system have relatively high practicability and expandability.

The present application relates to the technical field of data management and intelligent analysis of kinetic simulation engineering, and particularly relates to a structured intelligent archiving review method and system for kinetic simulation engineering. The input of the method is a target kinetic simulation engineering catalog and its configuration file, running artifacts, result artifacts and document basis; the processing process includes engineering scanning, case identification, specification input selection, configuration rule review, running health assessment, structured evidence organization, evidence constraint question and answer generation and report verification; the output is structured evidence, question and answer result, research report and audit record. Through the above technical chain, the present application realizes closed-loop processing from engineering data identification, evidence organization, question and answer generation to report verification, improves the kinetic simulation engineering data management efficiency, conclusion reliability and result traceability, forms a research report with complete evidence traceability, and comprehensively improves the simulation engineering data management efficiency, conclusion reliability and audit reliability.

The embodiment of the invention discloses a mining truck difficulty identification model training method, a difficulty identification method and equipment. The method comprises the following steps that collection data are obtained, a first sample set is generated according to the collection data, and the first sample set comprises normal cases and real difficult cases; performing difficult case expansion processing on the real difficult case to generate a simulation difficult case; summarizing the real hard cases and the simulation hard cases into hard cases, and summarizing the hard cases and the normal cases into a second sample set; and training the initial model by using the second sample set, and marking the trained initial model as a difficult case identification model. Therefore, according to the method and the device, the existing collected data can be utilized to simulate and generate more difficult cases, and the difficult cases in the original sample set are replaced or added, so that the expansion of the difficult cases is realized, and the quality of the sample data is improved. And subsequent model training can improve difficult case identification and multi-target perception capabilities, and is helpful for popularization of mine card automatic driving.

The invention relates to an unknown pneumonia risk case identification and information acquisition method, which comprises the following steps of: extracting clinical signs and symptoms on the basis of a large language model in combination with an RAG technology and an MCP technology, calculating a risk score by using a public health risk case identification rule model algorithm, and judging a risk case, and transmitting data in clinical and disease control application scenes through a plug-in and an informatization platform system, and reminding a worker to standardly collect risk case information. The problems of low accuracy and difficult practical application in the aspect of pneumonia risk case identification by using electronic medical record data are solved, the identification accuracy of unknown reason pneumonia risk cases is greatly improved, missed diagnosis and misdiagnosis conditions are reduced, a large amount of electronic medical record data can be quickly processed, the integrity and accuracy of collected information are ensured, and the method is suitable for popularization and application. The pneumonia type of a case and the potential epidemic risk can be quickly judged.

The application discloses an intelligent litigation case identification method and system based on RAG and a knowledge graph, relates to the technical field of litigation case identification, and comprises the following steps: performing semantic vectorization processing on litigation case texts based on a pre-trained text embedding model, and outputting a semantic vector sequence; performing similarity matching on the semantic vector sequence and obtained knowledge graph entity vectors, obtaining related graph entities, and constructing context corpus blocks corresponding to the litigation case texts; inputting the context corpus blocks into a large language model, identifying elements of the litigation case through structured prompt words, and outputting structured case information. The application is used to solve the problems of low entity recognition accuracy and disconnection of knowledge retrieval results in the prior art.

A method and system for process mapping including: a processor; and a memory storage operatively connected with the processor, wherein the memory stores computer-readable instructions, which, when executed by the processor, performs the method of: recording activities performed by employees on a plurality of computers, wherein the activities comprise screens used and related data that can be obtained through a web-based connection; analyzing the screens used and related data to generate employee activity data; determining processes based on the activity data; and generating a process map based on the determined processes.

The invention discloses a handwriting experiment sample intelligent extraction system, and relates to the field of material evidence investigation and identification, and the system comprises an information collection module which is used for collecting related data and generating structured data; the related data comprises personnel information, case information and environment data; the handwriting generation module is used for processing the structured data and generating handwriting content meeting the identification requirement; the handwriting content comprises a handwriting text and a voice file; and the sample confirmation module is used for carrying out integrity confirmation on the handwriting content and carrying out authenticity verification on the extraction person and the extracted person to generate a confirmation result. The handwriting experiment sample extracted by the method can meet the identification requirement, and the case identification quality and efficiency are improved.

The embodiment of the invention relates to a case recognition model training method and device, electronic equipment and a storage medium. The method comprises the steps that a first data set is acquired; identifying each piece of first sample case data in the first data set through a plurality of first case identification models, so as to determine first identification accuracy rates corresponding to the plurality of first case identification models; obtaining a second data set corresponding to at least one second case identification model; and according to multiple pieces of second sample case data contained in a second data set corresponding to each second case identification model, training each second case identification model to obtain a target case identification model corresponding to each second case identification model. According to the case recognition model training method and device, the electronic equipment and the storage medium, the current medical scene can be adapted more quickly and accurately.

The invention discloses an expert recommendation system for multi-department coordinated consultation of lung difficult miscellaneous diseases, and belongs to the technical field of medical monitoring and artificial intelligence. The expert recommendation system comprises a data processing unit, a patient portrait construction unit, an expert information management unit, an expert recommendation unit, a system implementation and application unit and a medical large model auxiliary diagnosis unit. The patient portrait construction unit is used for establishing a patient portrait and automatically updating the patient portrait through an incremental learning mechanism; the expert information management unit is used for collecting expert information, evaluating expert ability by utilizing an expert evaluation system and maintaining an expert database; the expert recommendation unit is used for performing expert recommendation based on an expert recommendation algorithm; the system realization and application unit is used for realizing an interaction function, a background processing function and a decision support function of a front-end interface; and the medical large model auxiliary diagnosis unit is used for executing multi-modal data fusion, disease probability prediction and difficult case identification through a deep learning model.