188results about How to "Improve professionalism" patented technology

The invention discloses a generation and evaluation method for acquiring long text description from a laparoscopic surgery image, which comprises the following steps of: constructing a cue word containing the laparoscopic surgery image, short text description, a target bounding box and surgery knowledge, and constructing long text description data by utilizing a large language model based on the cue word; constructing an image long text description generation model which comprises a pre-trained visual encoder, a visual feature query converter, a language decoder and an external operation concept vector database; performing supervised training on the image long text description generation model by using the long text description data; constructing a long text description evaluation method to evaluate the trained model; the method comprises the following steps: acquiring an input laparoscope image video stream through a laparoscope image system, and screening a key frame as a to-be-processed image; and inputting a to-be-processed image into the trained and evaluated model to obtain an image long text description result. According to the method, the ability of understanding operation details in the image can be enhanced, and more comprehensive operation image description can be generated.

The invention discloses an intelligent financial statement automatic generation platform, and relates to the technical field of financial statement generation, and the platform comprises a data integration module which is used for connecting and synchronizing a data source, the data source comprises a bank transaction record, an enterprise resource planning system and an electronic invoice system, and data is grabbed through connection configuration of a data interface. According to the intelligent financial statement automatic generation platform, classification rules of financial subjects are customized by a user, so that the platform can adapt to specific financial statement requirements and business characteristics of different enterprises, the flexibility and applicability of the system are improved, statement customization is supported through the platform, financial statements of different versions are provided, and the user experience is improved. According to the method, different requirements of different users on financial information are met, the flexibility and applicability of the system are improved, a temporary version exclusive to the current requirement is formulated based on a customizable mechanism of a classification rule, and more targeted financial support is provided for decision and management of enterprises.

The invention relates to an intelligent semantic analysis knowledge graph system for multi-source heterogeneous data. The system comprises a data preprocessing module used for preprocessing to-be-processed multi-source heterogeneous data to obtain format-standardized text data; the semantic analysis module is used for performing natural language processing on the text data to obtain words, and then performing vocabulary layer, syntactic layer and semantic layer analysis on the text data and the words to obtain structured data; the structured data comprises a triple consisting of entities, relationships and events and a network diagram of the entities and the relationships; the model structure mapping module is used for mapping the structured data into a predefined structure of a knowledge graph to obtain structured graph data; and the knowledge graph module analyzes entities and relationships in the graph data, performs reasoning to obtain potential relationships in the graph data by utilizing a logic rule and a machine learning algorithm, realizes cross-data-source and cross-document deep semantic association, and improves the efficiency of large-scale data processing and knowledge management.

The invention discloses a medical field large model pre-training and fine tuning method and device based on a knowledge graph, and the method comprises the steps: training a reward model through employing collected medical related text data and a quality marking result, and screening out high-quality data from the newly obtained medical related text data through employing the trained reward model; coding the medical knowledge graph into a text format processed by a large language model, and alternately inputting the high-quality data and the coded medical knowledge graph into the large language model in different batches for model pre-training; and designing an instruction fine tuning data format comprising an instruction, a knowledge graph reasoning path and a generated answer, inputting a training sample constructed in the instruction fine tuning data format into the pre-trained large language model for model fine tuning, and outputting a final answer and reasoning path according to the instruction newly input by the user by utilizing the fine-tuned large language model. According to the method, the professionality of the large language model in the medical field can be effectively improved, and AI application in high-risk fields such as medical treatment is promoted.

The invention provides a medical AI question and answer method based on retrieval enhancement, which comprises the following steps: acquiring a medical professional knowledge base, and constructing an entity recognition model and a medical knowledge graph; the method comprises the following steps: splitting a medical professional knowledge base, and creating a vector index and a full-text retrieval reverse index; intention classification is carried out, and initial processing is completed; performing multi-path mixed retrieval on the medical professional knowledge base to obtain a user question related text block; a filtering rule is preset, and the user question related text blocks are filtered; answering the user question according to a preset second prompt word, and answering the user question according to a preset third prompt word when the user question exceeds the range of the medical professional knowledge base or the multi-path mixed retrieval has no result. According to the method, related context information can be accurately searched, reliable reference information is provided for a generative large model to answer professional questions, the interactive experience of a medical question answering system is improved, and the accuracy and specialty of answer generation of the system are remarkably improved.

The invention relates to a text semantic conflict detection method for constructing context association based on a large model. The method comprises the following steps: S1, segmenting an original text by using a segmentation function; s2, performing triple information extraction on each segmented text by using a large model to generate a data set of an entity corresponding relationship; s3, extracting data groups of all entity pairs and relationships, and performing confidence evaluation; s4, aggregating all triads, constructing a affair graph, and pointing out a contradictory relationship with relatively low confidence in the affair graph; and S5, for the found contradictory relationship, performing semantic conflict repair in the text in combination with the large model. According to the method, the accuracy of information generation of the large language model is remarkably improved in a context association mode.

The invention particularly relates to a document abstracting method based on domain knowledge and a multi-granularity graph network, which comprises the following steps: S1, constructing a knowledge coding module, carrying out domain knowledge graph transfer learning, extracting triple structure features in combination with a capsule network, and generating a text entity coding module; s2, constructing a knowledge fusion module, converting a document text into a semantic vector, and adaptively fusing the semantic vector with external knowledge; s3, constructing a word-level semantic graph, and aggregating neighborhood information through a semantic graph attention mechanism to obtain word embedding representation containing context; s4, constructing a sentence-level semantic graph, performing self-attention calculation on the word-level semantic vector in the S3 to obtain semantic node representation of the sentence-level semantic vector, and performing feature aggregation on sentence-level nodes to obtain global semantic information of the document; and S5, constructing an output layer, and predicting according to the document feature vector in the S4 to obtain a final extraction type abstract result. The method has the advantages that redundant information in the document is removed, and ambiguity of text semantics is eliminated.

The invention discloses an intra-domain model training method and system based on a knowledge graph, and belongs to the technical field of artificial intelligence, and the method comprises the steps: constructing a domain knowledge graph; extracting newly added high-frequency professional terms from the domain knowledge graph; generating a newly-added lexical element for each newly-added high-frequency terminology, and integrating the newly-added lexical element into a vocabulary of the pre-trained large language model to obtain an expanded vocabulary; initializing a word embedding vector of the lexical element by utilizing an entity embedding vector in the knowledge graph, adjusting a word embedding layer of the pre-training large language model based on the word embedding vector of the newly-added lexical element in the expanded word list, and carrying out retraining; taking question and answer pairs constructed based on the domain knowledge graph as an instruction fine tuning data set after being audited and perfected by experts, and performing fine tuning on the retrained large language model; and performing model performance evaluation and continuous optimization. The method can provide a more comprehensive word list, improve the understanding ability of the model for vocabularies in a specific field, improve the model training efficiency, and is suitable for large language model training in the field.

The invention relates to the field of artificial intelligence, in particular to an answer matching method and system for an artificial intelligence large model, and the method comprises the steps: obtaining a question text of the nth question asked by a user, carrying out the feature extraction of the question text through a natural language processing technology, and constructing a question feature set; inputting the question feature set into a difficulty evaluation model, and outputting a difficulty score of the question text; based on the difficulty score of the question text, performing matching degree calculation on all the artificial intelligence large models in the large model library by adopting a large model matching degree formula to obtain the matching degree of each artificial intelligence large model; and analyzing the matching degree of each artificial intelligence large model according to a preset large model matching strategy to obtain a selection scheme. By adopting the method, the matching degree between the question and various large models can be accurately and quantitatively measured, so that the professionalism and accuracy of question answering are improved; and meanwhile, the continuous dialogue processing capability of the large model can be improved.

The invention discloses an artificial intelligence automatic report evaluation method and system, and the method comprises the steps: carrying out the focus information extraction of a to-be-evaluated text report outputted by a medical artificial intelligence model and a pre-established gold standard text report through a trained large language model, and converting the focus information into structured information; and obtaining a to-be-evaluated text structured report and a gold standard text structured report. In order to solve the problem that the lesions between the two reports need to establish a corresponding relationship for evaluation, the method further introduces a structured report matching algorithm, so that a corresponding matching relationship is established for the lesions in the two reports. On the basis, whether the attributes of the single focus are consistent or not can be evaluated, so that the performance of the text report is subjected to multi-level quantification.

The invention discloses a document proofreading processing method based on a large model. The document proofreading processing method comprises the following steps: step 1, preprocessing a document and identifying the document as a text; step 2, segmenting the text into independent text blocks; step 3, performing vectorization processing on the text blocks; step 4, performing retouching on the text block by using the random degree large model to obtain a new text block; step 5, performing vectorization processing on the new text block; 6, comparing the vector similarity of the text block and the new text block, if the similarity is high, entering the next step, and if the similarity is low, repeating the steps 4-6; 7, comparing the text of the text block with the text of the new text block, and marking difference characters in pairs; and 8, proofreading the text blocks and the new text blocks according to the document, and integrating and outputting the text blocks and the new text blocks. According to the method, more accurate and comprehensive document proofreading service can be provided, dynamic change of languages and diversified document requirements can be met, and the specialty and credibility of documents are remarkably improved.

The invention relates to the technical field of deep learning, and discloses a potential hot content identification method based on a mainstream value droop domain knowledge base, which comprises the following steps: constructing the mainstream value droop domain knowledge base, carrying out topic modeling through a hidden Dirichlet distribution model and extracting potential topic words, an event triple is extracted through a RoBERTa-CRF (Conditional Random Field) model, and an event atlas is constructed; nodes related to the hot spots are retrieved from the event atlas through an atlas retrieval engine, sub-graphs formed by the related nodes are converted into vector representation through a graph embedding technology, and a hot spot mode is formed; and by calculating Jaccard similarity, matching content features similar to the hotspot mode in the fusion graph representation network, and identifying potential hotspot content. According to the method, the forward enhanced information content is accurately identified and screened through thematic modeling, event extraction, atlas retrieval, similarity calculation and other technologies, and the real-time monitoring and prediction capability on the social hotspot trend can be enhanced.

The invention discloses a nursing information distribution management system based on big data, and relates to the technical field of information distribution management, and the system comprises a data collection module which is used for collecting original data from a nursing data source in real time and cleaning the original data, the original data comprises patient information, nursing records, resource use conditions and nursing personnel capabilities; the data acquisition module is connected with an external system through an API (Application Program Interface); the nursing information distribution management system based on big data can comprehensively evaluate the skill level and professional ability of nursing personnel, update skill archives, and provide an accurate basis for personnel deployment; appropriate nursing personnel are allocated to appropriate tasks, so that the quality and the specialty of nursing services are improved; optimal allocation of resources and nursing personnel is realized; resource preparation is made in advance, resource waste and shortage are avoided, and the resource utilization rate is increased.

The invention discloses a medicine packaging box pattern printing defect detection method, and relates to the technical field of printing quality detection, and the main scheme is as follows: collecting to-be-detected printing data and standard printing packaging box template data, and then calculating a character deviation index, a pattern position coordinate index and a to-be-detected color index according to the data; wherein the character deviation index is obtained by the pixel and format similarity index of the character area; the pattern position coordinate index is calculated based on the standard position coordinate system number and the data to be measured; a color index to be detected is calculated by a deviation color coefficient and the like, then text, position and color index thresholds are set, whether a text area, a pattern position and a pattern color are qualified or not is judged by comparing each index with a corresponding threshold, the printing defect is judged to exist as long as one of the indexes is unqualified, and the method can effectively detect the printing defect.

The invention discloses an industrial measurement data intelligent analysis report generation method and system based on a large language model, and relates to the technical field of industrial measurement, and the method comprises a multi-source data and abstraction module which carries out the structural processing of original data DAT output by measurement analysis software; and the large language model service and interface module is used for submitting the PCT to a selected large language model LLM through a multi-model adaptation interface to execute semantic reasoning and output a structured text TXT. Structured analysis and semantic association modeling are carried out on original measurement data through multi-source data and an abstract module, when overall alignment deviation or local feature anomaly exists in the original data, a hierarchical dependency relationship between the data can be established through a semantic graph structure, a deviation transmission path is revealed, and the accuracy of the measurement data is improved. The large language model is analyzed in a unified data context, and the utilization depth of measurement data can be effectively improved, so that the accuracy and integrity of report analysis are improved.

The invention discloses a method and system for generating related cue words of science and technology texts based on a large model, and the method comprises the steps: analyzing simple cue words input by a user, determining technical subject words, background words and key words from the simple cue words, and further determining technical subjects, backgrounds and key points of the science and technology texts desired by the user; according to current simple cue words, science and technology cue words including newest technical branches, innovation concepts, potential problems and improvement effects are derived to further improve the cue words, and science and technology texts are generated after the improved cue words are input into the second large model to achieve the purposes of improving the specialty and the information amount of the texts and improving the efficiency of the science and technology texts. And the high-standard writing requirement in the science and technology field is met.

The invention discloses a protection setting value intelligent comparison method and system, and relates to the technical field of power protection setting value check, and the method comprises the steps: carrying out the first analysis of a file format, and storing the file format in a database; acquiring real-time data, and performing first comparison on the data; and visualizing the comparison result and automatically generating a check report. According to the protection setting value intelligent comparison method provided by the invention, the electronic setting value sheets in various formats are automatically imported and analyzed, and the data are stored in a standardized manner, so that the processing efficiency and accuracy of the setting value sheet data are improved, the operation setting value data of the protection device are obtained in real time and are integrated into a setting value checking system database, and the safety of the protection device is improved. According to the method, the synchronization of the operation data and the scheduling data is improved, the scheduling issuing fixed value and the operation fixed value are compared item by item by adopting an automatic algorithm, the efficiency and the accuracy of the fixed value checking work are improved, and better effects are achieved in the aspects of accuracy, synchronism and efficiency.

The invention relates to the technical field of artificial intelligence and legal information processing, in particular to a legal consultation intelligent response method based on deep learning. The method comprises the steps that data preprocessing is conducted on legal consultation questions input by a user, and a semantic nested matrix is generated; dynamically adjusting a nesting weight, and constructing a generation matrix in combination with the semantic nesting matrix; comparing the generated matrix with a legal knowledge base matrix layer by layer to confirm the matching degree of the generated solution and the elements of the legal knowledge base matrix; taking a result with the highest matching degree as a preliminarily generated solution, and generating a feedback matrix; and optimizing and updating the generated matrix based on the feedback matrix to obtain a generated solution matrix. The problems that an existing intelligent legal consultation system is insufficient in legal semantic understanding, limited in semantic weight adjustment, poor in logicality of generated content, low in legal clause matching degree, lack of a self-adaptive optimization mechanism fed back by a user and incomplete in expression ability of natural language output are solved.

The invention discloses an intelligent data retrieval method based on LLM, and relates to the field of application of large models. Comprising the steps of 1, respectively constructing a text knowledge base and a data resource knowledge base for guiding the analysis direction of a large model, reducing the answer range of the large model and guiding the large model to generate answers, and 2, performing context association on incomplete semantics of a question by using a large model LLM according to the question input by a user to obtain a user question with context information, 3, performing intention recognition and task analysis on the user problem with the context information by using the large model, and determining information required to be queried by the user; 4, performing semantic similarity retrieval in the data resource knowledge base according to a task of large model analysis; and 5, post-processing the retrieved data resources, performing streaming output on a processed result, and performing front-end display.

The invention provides an engineering investigation report automatic generation system and method based on a BERT model. The system comprises a data preprocessing module, a BERT model optimization customization module, an information extraction and structuring module and a report generation engine. According to the method, the limitation that time and labor are consumed and errors are prone to occurring in the traditional report compiling process can be overcome, the engineering investigation data can be automatically analyzed and processed by integrating an advanced BERT model for optimization and customization, then a high-quality report meeting the industry standard and the customer requirement is generated, and intelligent processing of the engineering investigation data and automatic report compiling are achieved.

The invention relates to a retrieval enhancement method, device and equipment based on domain term enhancement and a storage medium, which are applied to the field of natural language processing, and the method comprises the following steps: respectively extracting specific domain knowledge and other domain knowledge in a domain knowledge base; converting the specific domain knowledge and other domain knowledge into vectors and storing the vectors in a vector database; training the large language model to obtain a retrieval mark prediction model; when a query statement is received, predicting a retrieval mark type # imgabs0 # by using the trained retrieval mark prediction model; and based on the retrieval mark type # imgabs1 #, making an answer corresponding to the query statement by using the large language model and specific domain knowledge and other domain knowledge stored in the vector data. The method has the technical effects that professional answers can be specifically made based on the extracted knowledge in the specific field, the possibility of misunderstanding or ignoring important information in the retrieval process is reduced, and the accuracy of document retrieval and the precision of retrieval answers are improved.

The invention provides a DCS electrical isolation device set diagram design method and device, and the method comprises the steps: obtaining an electrical isolation signal list, and constructing an electrical isolation signal configuration table according to all signals of various types in the electrical isolation signal list; defining macro elements of the electrical isolation device based on the signal information and a standard drawing template to obtain a plurality of graph macros; defining a break point name according to the path of the target signal in the electrical isolation signal configuration table; and designing an electrical isolation device complete set diagram based on the electrical isolation signal configuration table, the plurality of diagram macros and the interruption point names by utilizing preset electrical design software. According to the method, the electrical isolation signal configuration table, the electrical isolation graph macro and the interruption point are defined in detail by adopting a parameterization design method, and automatic drawing is realized. And by standardizing the design sequence, the risk of human errors is reduced. And the design method and steps are kept consistent, so that the consistency of design quality is ensured, and the standardization and specialty of the overall design are improved.

The invention discloses an LLM (Logical Language Modelica) domain method for Modelica intelligent modeling optimization, and belongs to the field of computer-aided modeling. Structured information is converted and extracted by collecting field data of the new energy automobile; then, based on the RAG technology, intelligent blocking and vectorization processing is carried out on the structured information to form a structured knowledge base, and then a recall test is carried out by calculating a comprehensive weighted score to judge whether the structured knowledge base is available or not; then, the LLM is integrated into an AI-Agent framework, an AI-Agent is formed, and LLM parameters are configured; integrating the structured knowledge base which passes the test into the AI-Agent by utilizing an API (Application Program Interface) and an access protocol; and meanwhile, a specialized cue word template is designed and integrated into the AI-Agent, and a Modelica code generation template and related constraints are standardized. And finally, the user asks a question to the AI-Agent, the LLM works according to a specialized cue word instruction, an API interface is called to access the knowledge base, and Modelica codes which conform to language specifications and can be operated by the user are automatically output through the LLM. According to the invention, end-to-end conversion from unstructured input to high-fidelity model output is realized.

The invention relates to the technical field of medical diagnosis accompanying services, and discloses an order matching calculation method for diagnosis accompanying services, which is used for optimizing the distribution and matching process of the diagnosis accompanying services. The method comprises the following steps of setting patient information, when the patient information is generated, marking the patient information as an accompanying diagnosis event, enabling one piece of patient information to correspond to one accompanying diagnosis event, collecting disease condition keywords, obtaining patient disease condition filling content information in the accompanying diagnosis event, and collecting disease condition keywords in the patient disease condition filling content information, the method comprises the steps of forming a set, setting an illness condition keyword library and danger degrees, executing a first person screening strategy according to an illness condition danger degree value of a patient and accompanying diagnosis person information in an accompanying diagnosis person set, generating a first target person set, executing a first order matching strategy according to the first target person set, and generating an accompanying diagnosis service order. And when the diagnosis accompanying service cycle corresponding to the diagnosis accompanying service order is finished, executing a personnel score optimization strategy.

The invention relates to a multi-agent cooperation-based language model reasoning resource scheduling method and system, and the method achieves the intelligent simulation of multi-view iterative thinking in a complex reasoning task through the construction of a multi-agent system which is clear in division of labor and is provided with a special knowledge base. View limitation and decision deviation of a single model in long-sequence and multi-step reasoning are effectively overcome; depending on the fusion of the general capability of the large language model and the task special knowledge base, the professionality and accuracy of the reasoning result are improved. Meanwhile, the problems of resource waste, redundant calculation and unstable convergence caused by cognitive overload in the cooperation process are solved by combining multi-round dynamic cooperation with a convergence mechanism of cognitive load perception and a dynamic token number limitation and low-confidence branch pruning strategy; therefore, on the premise that the reasoning quality is guaranteed, the utilization efficiency of computing resources is remarkably optimized, peak value occupation is reduced, the overall reasoning time delay is shortened, and reliable technical support is provided for efficient and stable deployment of a large language model in a complex task.

The invention provides an emergency plan auditing method and system, electronic equipment and a storage medium. An emergency plan is analyzed to obtain plan content; performing text cleaning and formatting on the chapter content of each chapter, the paragraph content of each paragraph, the entry content of each entry and a plurality of titles which are obtained by performing document segmentation on the emergency plan according to the plan content and dividing according to the logic hierarchy of the emergency plan; obtaining a target chapter content, a target paragraph content, a target entry content and a target title; extracting a plurality of keywords related to the emergency field from each target chapter content, each target paragraph content and each target entry content, and endowing each chapter, each paragraph and each entry with a corresponding semantic category; and adding field information for each chapter, each paragraph and each entry, and constructing a nested structure for each chapter, each paragraph and each entry added with the field information according to the logic level of the emergency plan to obtain a corresponding tree data model.

The invention provides an intelligent teaching material generation method, device, equipment, product and system based on a problem chain, and the method comprises the steps: obtaining teaching task information which comprises the course content of a target course and a teaching target; calling a pre-trained large language model to generate a problem chain of the target course based on the teaching task information, and generating a teaching design of the target course based on the problem chain; wherein the question chain comprises a plurality of questions which are used for guiding students to learn the course content and achieve the teaching target and have the association relationship. The problem is intelligently generated through the large language model to solve the guided teaching design, the students can be guided to gradually and deeply learn, the learning interest of the students is stimulated, the teaching process is more orderly and coherent, the students are helped to form a systematic knowledge framework, and the learning effect is improved.

The embodiment of the invention provides a voice interaction method and device, electronic equipment and a medium, and belongs to the technical field of artificial intelligence. The method comprises: receiving voice data of a target user, and converting the voice data into text data, the target user being a user belonging to a target user group; according to the text data, at least one associated question and answer statement data is determined in a question and answer knowledge base, and the question and answer knowledge base comprises question and answer statement data for the target user group; generating a question-answer instruction by using the at least one associated question-answer statement data and the text data; inputting the question-answer instruction into a language question-answer model, so that the language question-answer model determines answer text data in a question-answer knowledge base or at least one associated question-answer statement data according to the question-answer instruction; and converting the response text data into response voice data and outputting the response voice data. Therefore, the accuracy and specialty of voice interaction can be improved, and the defect that professional field information cannot be provided in the prior art is overcome.

The invention discloses a bidding and tendering file generation method and system, and the method specifically comprises the following steps: S1, bidding file analysis: carrying out the structural analysis of a bidding file, and extracting key information; s2, a template selection and calling step: selecting a proper bidding file template according to the type of the bidding file and the item category; the invention relates to the technical field of data processing. According to the bidding and tendering file generation method and system, the bidding file is subjected to structured analysis in an automatic mode, and the tendering file meeting the requirement is quickly generated. Therefore, the time and the labor cost for manually compiling the bidding document are greatly reduced, an enterprise can respond to the bidding demand more quickly, the possibility of human errors is reduced due to automatic generation, and the accuracy and the consistency of the document are improved.

The application provides a multi-agent collaborative management method and system for whole-process engineering consultation, which comprises the following steps: S1, obtaining a whole-process engineering consultation task; S2, performing task decomposition on the obtained whole-process engineering consultation task based on an LLM large language model to obtain task decomposition features of the whole-process engineering consultation task, wherein the task decomposition features comprise one or more subtasks; S3, respectively performing task classification feature extraction on each subtask of the task decomposition features to obtain a task classification label corresponding to each subtask; S4, according to the task classification label corresponding to each subtask, matching a corresponding agent from candidate agents to process the subtask, and obtaining a subtask solution output by the agent; and S5, performing optimization processing according to the obtained subtask solution to obtain a whole-process engineering consultation task solution. The application helps to improve the effect of the whole-process engineering consultation task solution.