563 results about "Collation" patented technology

The invention discloses an intelligent report generation method and system based on multi-source heterogeneous data fusion, and particularly relates to the technical field of data fusion, and the method comprises the steps: constructing a fact anchor point containing a stable identifier in a fusion layer, and solidifying an aperture version signature and a time interval; packaging a screening condition, a sorting rule, an access plan, time window limitation and the like into a controlled context packet and freezing the controlled context packet into a unique fact source; in the range, generating sentences by using a limited template and writing sentence-level reference marks, so that each sentence can be traced back to a data slice and an evidence fragment; performing consistency verification according to a source, an aperture and time, performing conservative degradation on the attribution expression according to a consistency threshold value, and outputting a degradation strategy description; and finally, page two-way pointer and cross-sentence consistency recording and audit playback are realized through an evidence chain pointer table and a published version signature, and the problems of caliber drift, cross-window access and difficulty in evidence tracing are solved.

Systems and methods disclosed herein are enabled to dynamically generate structured documents using one or more artificial intelligence models. A computing device receives an output generation request and uses a first AI model to retrieve data chunks from source documents and applicable templates. A second AI model ranks the retrieved chunks based on one or more metrics, such as vector similarity, keyword density, and temporal relevance. A third AI model subsequently generates a response using the ranked chunks, templates, and predefined operational boundaries for each chunk. The generated response is tagged with source identifiers to enable the traceability of the response back to corresponding chunks. The system transmits, via the computing device, the response, the retrieved chunks, and / or the source identifiers.

The invention provides a household appliance knowledge question-answering method and system based on retrieval enhancement generation. The method comprises the following steps: acquiring household appliance field multi-modal data from a multi-format document library; extracting text information, table information and chart information in the multi-modal data; performing domain term injection processing on the extracted information, and constructing a packet domain enhancement index; receiving a natural language question input by a user; the natural language problem is analyzed through a query optimizer, and semantic retrieval and keyword retrieval are executed in parallel; carrying out fusion processing on the semantic retrieval result and the keyword retrieval result; selecting matched document fragments by adopting a relevancy sorting algorithm; inputting the matched document fragments into a large language model to generate candidate answers; verifying the compliance and traceability of the candidate answers through a credibility evaluation module; outputting a final answer with a reference source; and storing the high-frequency questions and the final answers into a cache library to solve the problems that the answer accuracy of a knowledge question-answering system is reduced and the response efficiency is limited.

The invention provides a method and system based on NLP file analysis, and relates to the technical field of natural language processing. According to the method, time and identifier unification and format and character set standardization are carried out on the multi-source file, layout segmentation, table structure extraction, reference analysis, term standardization and anaphora resolution are combined, semantic representation is constructed, a hierarchical index and a unique traceability identifier are generated, intention recognition, retrieval sorting, incremental updating and consistency verification are supported, and the method is suitable for large-scale popularization and application. Unification, semantization and traceability of the file analysis process are achieved, and the processing efficiency and accuracy are improved.

The invention relates to the technical field of language information processing, and provides a large language model-based structured information retrieval method, which comprises the following steps of: deploying an adapter, accessing operation and maintenance data, carrying out timestamp synchronization on an original entry and identifying a source, mapping equipment, parts and personnel based on an asset directory and recording mapping confidence, extracting structural elements from the records; injecting three types of metadata into candidate evidences, establishing corresponding nodes and edges in a graph database, analyzing query into a structured retrieval intention, expanding a candidate evidence set along three chains to form candidate sub-graphs, and calculating comprehensive scores for sorting; the candidate evidences should be verified and constrained, meanwhile, LLM is called for inference, feasibility inference confidence is output, actual measurement results, the inference confidence and source confidence are fused according to preset weights, feasibility scores of the candidate evidences are obtained, and for the candidate evidences passing feasibility verification, accurate anchor points are calibrated for each evidence along a source chain; and generating a structured answer.

The invention discloses an analyzing and editing system and device for high-frame-rate rendering of a PDF document, and the system comprises a file list uploading module which is used for uploading a pdf file and displaying a version historical record list and information; the document analysis logic module is used for reading and analyzing document information, determining a block level sequence of obtained data, deleting useless elements, and carrying out content sorting and splicing according to a layout; the pdf original document display module is used for executing pdf document title display operation, document display page cutting operation, document directory generation operation and document analysis display result operation in original document uploading, and the document directory generation operation comprises the steps of capturing multi-level digital numbers of documents, classifying number types according to matching results of capture groups in regular expressions, and storing the classified numbers in a database; and then numbering processing is carried out through a specified hierarchy mapping rule, and finally a file directory number is obtained. The problems that a traditional tool is inaccurate in analysis, low in editing efficiency and poor in professional adaptation can be effectively solved.

The invention relates to an intelligent document information real-time retrieval method and system based on the RAG technology, and the method comprises the steps: analyzing documents of various formats, extracting a text, maintaining the content continuity through adaptive semantic partitioning processing, and building a character-level position index at the same time; after vectorizing the text blocks, generating a plurality of rewriting queries for the original query; respectively carrying out mixed retrieval (combining keywords and semantic retrieval) for each rewriting query, and fusing and reordering results to obtain candidate text blocks; after correlation filtering, inputting a large language model according to correlation to generate an answer, and if the result is negative, triggering secondary retrieval and reordering; and finally, outputting a structured answer containing position information and supporting front-end visualization. According to the method, the semantic integrity maintenance, the multi-format document processing efficiency and the key information retrieval accuracy are effectively improved.

The embodiment of the invention relates to the technical field of data processing, and particularly provides a smart library information retrieval method and system based on AI session interaction. According to the embodiment of the invention, an information retrieval session request initiated by a user through a smart library interaction interface is received; collecting session interaction data containing natural language query content and context association information; performing semantic structuring processing on the session interaction data to generate a session semantic feature set containing an entity relationship network and an intention classification vector; carrying out deep mining on the session semantic feature set by utilizing a debugged retrieval intention understanding model, identifying potential retrieval demands and demand priority ranking, and generating retrieval demand feature vectors; and on the basis of library collection information in the vector matching intelligent library resource database, generating an information retrieval result set containing resource association degree sorting, and feeding back the information retrieval result set to the interactive interface for visual display. According to the embodiment of the invention, the accuracy and comprehensiveness of information retrieval are improved, and more intelligent and efficient retrieval experience is provided for users.

The invention provides an efficient annotation-driven hierarchical fault positioning method, which comprises the following steps of: guiding a large language model to intelligently analyze and generate semantic annotations of a text based on a positioning algorithm of large model annotations, and performing fine-grained software fault positioning from a file level to a function level and then to a position level by using the annotation-based hierarchical positioning method. And an efficient hierarchical progressive sorting and screening algorithm is used for ensuring the software fault positioning efficiency. According to the method, the defect positioning efficiency can be ensured while the positioning accuracy is ensured. The positioning algorithm based on large model annotation strategically balances the information density, provides enough context clues, and improves the model understanding ability; according to the annotation-based hierarchical positioning method, positioning is divided into file / function / position hierarchies, and annotations are added by using a large model cue word project in sequence, so that the positioning accuracy is improved; and an efficient hierarchical progressive sorting and screening algorithm is provided, so that the optimal balance between the performance and the cost is realized.

The invention relates to the technical field of building information, and provides a knowledge base construction and retrieval method and system based on a multi-source text in the building field, and the method comprises the following steps: a knowledge base construction stage: constructing a multi-dimensional metadata feature vector for a multivariate text based on a standard classification index table; the method comprises the following steps of: converting and segmenting a document, splicing an end clause with all superior title texts by utilizing a context inheritance algorithm to form a text unit with complete semantics, and performing dynamic filtering based on an analyzed query intention and a metadata vector at a user retrieval stage; then, in the screening set, performing fusion calculation on semantic vector similarity, keyword matching degree and authority offset weight based on effectiveness attribute and implementation time, and performing mixed retrieval and reordering on the text units; and finally selecting a text unit according to a sorting result and inputting the text unit into the large language model to generate answers. According to the method, high-precision and high-compliance intelligent retrieval and question answering of building domain knowledge are realized.

The embodiment of the invention discloses a multi-dimensional intention recognition method and device, computer equipment and a readable storage medium. The method comprises the steps of obtaining domain knowledge data, RAG data of user history retrieval and user portrait data, obtaining an original data set, converting the original data set into structured fields, and constructing a domain knowledge base based on all the structured fields; receiving a current question, preprocessing the current question to obtain a standard text, and performing semantic deep analysis on the standard text through a preset large language model to obtain a semantic analysis result; multi-dimensional scoring is carried out based on the semantic analysis result, weighted summation is carried out based on the multi-dimensional scoring result, and a plurality of candidate intentions and corresponding confidence coefficients are obtained; and performing descending sorting according to the confidence, and selecting the candidate intention with the highest confidence as an intention recognition result. According to the method, the recognition accuracy of the core appeal in the user question is improved.

The invention relates to the technical field of artificial intelligence, in particular to a question and answer method and system based on LLM fusion knowledge graph and retrieval enhancement generation. The entities which are connected with the known entities and do not appear in the local knowledge graph are marked as related entities through the graph structure connection relation, a new local knowledge graph is formed, descending sorting is carried out according to the edges and the weights in the connected components, when the weights are the same, the lexicographical order of the entity names is used as a second sorting key, and the second sorting key is used as a third sorting key; and after sorting is completed, according to the sorting sequence, until the accumulated number of the documents reaches a preset number, combining the finally obtained individual documents with the questions, and inputting the combined documents and the questions into a large language model to generate a final answer. According to the method, the accuracy and interpretability of questions and answers can be effectively improved, the illusion problem of a large language model is solved, and reliable knowledge service support is provided for a complex decision-making scene.

The invention discloses an RAG question and answer optimization method and system based on fusion retrieval, and relates to the technical field of large model knowledge question and answer. The method comprises the steps of obtaining user query, executing keyword retrieval and semantic vector retrieval in parallel, and obtaining a keyword retrieval result list and a semantic vector retrieval result list respectively; and calculating the quality score of each text unit in each retrieval result list based on multi-dimensional quality evaluation, and filtering out the text units with the quality scores lower than a preset quality threshold value. And sorting the text units in all the filtered retrieval result lists by adopting a quality weighted reciprocal sorting fusion algorithm. And based on the length of each text unit and the document format to which the text unit belongs, performing context content enhancement and combination on each text unit after sorting, and generating an enhanced context block list. And constructing structured prompt information based on the enhanced context block list and user query, inputting structured prompt words into the large language model, and generating and outputting a final answer.

The invention belongs to the technical field of enterprise information technology resource management, and provides a tool performance evaluation method and device, and the method comprises the steps: carrying out the data preprocessing of a pre-constructed tool performance evaluation index pool matrix, and outputting a standardized matrix; n is the number of tools, and p is the number of selected indexes; performing PCA dimension reduction operation on the standardized matrix to obtain a principal component score matrix and a load matrix; d is the dimension after dimension reduction; performing k-means clustering on the principal component score matrix and the load matrix, outputting cluster division, and generating a clustering label for each cluster; and on the basis of an AHP analytic hierarchy process, tools in each cluster are sorted respectively. According to the method, the tools are grouped based on the clustering result through a clustering independent ranking mechanism, so that the intra-group homogeneity and inter-group difference of the tools are obvious, misjudgment of heterogeneous tools, monopoly sorting of dominant clusters and mismatching of resources are avoided, and the global ranking stability is improved.

The invention provides an RAG mixed retrieval method and device, and relates to the technical field of information retrieval, and the method comprises the following steps: obtaining a query text and a term library, and calculating to obtain a vector retrieval weight and a keyword retrieval weight according to the query text and the term library; obtaining a document library, and performing vector retrieval on the document library according to the query text and the vector retrieval weight to obtain a first candidate document set; performing keyword retrieval on the document library according to the query text and the keyword retrieval weight to obtain a second candidate document set; merging the first candidate document set and the second candidate document set to obtain a fused document set; and calculating, screening and sorting comprehensive feature scores of the documents in the fused document set to obtain a final document sequence, and inputting the final document sequence into the large language model to obtain an answer text. According to the method, the limitation of a single scoring dimension in professional field retrieval is overcome, and the correlation between the answer text and the query text is improved through multi-dimensional feature joint evaluation.

The invention relates to the technical field of data cleaning, in particular to an automatic data cleaning method and system based on a large language model.The method comprises the steps that original unstructured text data to be cleaned is preprocessed to generate standardized data blocks, and vector indexes are established for the data blocks; performing error detection on the data block based on a retrieval enhancement generation technology; in response to an error detection result, calling a large language model to generate a plurality of candidate repair schemes for the identified errors, and performing score sorting on the candidate repair schemes based on the multi-dimensional features; carrying out comprehensive confidence evaluation on the sorted candidate repair schemes, and executing a grading decision according to an evaluation result; and storing the result of the grading decision, and establishing a mapping relationship between the repair result and the original data to form a traceable data cleaning version record. Quantitative evaluation and three-level automatic decision making of the repairing result are achieved, and the problem that the result is uncontrollable due to the fact that LLM is directly used is solved.

The embodiment of the invention provides a question and answer data processing method and related equipment, and belongs to the technical field of natural language processing. The method comprises the following steps: carrying out hierarchical structure semantic division processing on a preset reference file, then carrying out semantic vectorization according to a preset text block obtained by division, and constructing to obtain text vector data; constructing a text vector retrieval model through a preset retrieval ranking algorithm according to the text vector data; splitting the preset user question through a preset large model to obtain a plurality of target sub-questions; according to the text vector retrieval model, performing retrieval sorting processing on each target sub-problem to obtain a target retrieval result; and generating a target question and answer result according to the target retrieval result and the preset user question. According to the embodiment of the invention, the understanding ability of the model for the user question can be effectively improved, and the accuracy of model answering is effectively improved.

The embodiment of the invention relates to the technical field of intelligent text analysis, in particular to a demand information duplication judgment method and device based on a large model, equipment, a storage medium and a product, and the method comprises the steps: converting a demand text uploaded by a user into structured demand information through structured processing; screening similar historical demands from a historical demand library based on vectorization retrieval and semantic reordering; character-level and semantic-level two-dimensional similarity calculation is adopted, and a comprehensive score is obtained through weighted fusion; and automatically judging a demand repetition state according to a score threshold, and synchronously outputting an interpretability conclusion containing functional intention similarity description and field coincidence degree analysis. According to the scheme, through double verification of large model semantic understanding and structured rules, the accuracy and interpretability of demand duplicate judgment are remarkably improved, and meanwhile, continuous optimization and self-adaptive extension of a historical demand library are supported.

The invention discloses an AI file query method based on human-computer interaction, which comprises the following steps: S1, collecting user input, and generating a query text; s2, intention recognition and slot extraction are performed on the query text, a slot set is generated, and standardization processing is performed; s3, utilizing the improved CoSent model to generate a query vector and a document vector, and storing the query vector and the document vector in a vector index; s4, retrieving the candidate document based on the query vector in the vector index, calculating semantic similarity and obtaining a comprehensive score in combination with a matching result; s5, carrying out reordering on the candidate documents; s6, generating a permission set based on the user identity, and executing permission filtering and field masking; and S7, inputting the document fragments passing the permission filtering and the query text into a text generation model, outputting a query result, and recording a log and user feedback. According to the invention, the accuracy and safety of file query are obviously improved.

The invention discloses a text matching positioning method, system and device and a medium. The method comprises the steps of obtaining an original long text and segmenting the original long text into ordered text blocks; selecting the first N text blocks to form a first candidate set based on the common substring sorting of the to-be-matched text and each text block; intercepting the first candidate set according to head and tail characters of the to-be-matched text to obtain a second candidate set; calculating similarity scores of the text blocks in the second candidate set and the to-be-matched text through a sequence matching similarity algorithm; combining the scores with the first candidate set to determine a matching result and output a position, and judging that no matching exists when all the scores are lower than a preset minimum threshold value. According to the method, the limitation of a traditional matching mode is broken through through primary screening of common substrings, accurate interception of head and tail features and multi-layer processing of a sequence matching similarity algorithm, complex scenes such as text missing and semantic equivalent expression can be effectively handled, the accuracy and robustness of text positioning are improved, and the user experience is improved. The method can be widely applied to the fields of document management, information retrieval, academic analysis and the like.

The invention discloses an auxiliary cadre recommendation method and system based on full-text-sparse-dense vector hybrid retrieval and big and small model collaboration, and the method comprises the steps: converting the original query of a user into a structured retrieval instruction based on a small-sample learning prompt project through a big language model; inputting the structured retrieval instruction into a trained post-post demand matching model, and analyzing and outputting a job condition set corresponding to the post; taking the job condition set as a retrieval condition, carrying out multi-dimensional parallel recall in the cadre information base through a triple mixed retrieval mechanism, and carrying out unified reordering on recall results to generate an ordered final cadre set; analyzing, optimizing and reasoning the final cadre set on the basis of a context prompt project and a large language model, and generating a final cadre recommendation scheme including man-post matching degree analysis; according to the method, intelligent and explainable cadre recommendation decisions can be realized, and the cadre selection efficiency, scientificity and objectivity are systematically improved.

The invention provides a document segmentation method and device based on a large language model, equipment and a storage medium, and relates to the technical field of text processing. The method comprises the steps of inputting a to-be-segmented target document into a pre-trained large language model, and executing the following operations through the large language model: performing text layout analysis on the target document, and identifying titles and all paragraphs of each level in the target document; for each paragraph, inserting an associated title related to the paragraph in all titles into an initial position of the paragraph to obtain a corresponding target paragraph; and sorting all the target paragraphs based on the semantic similarity among all the target paragraphs, and determining a segmentation result of the target document based on all the sorted target paragraphs. By the adoption of the technical scheme, when document segmentation is carried out, semantic loss in the document segmentation process can be effectively reduced, and therefore the document segmentation effect is improved.

The invention discloses an agricultural large model retrieval and evaluation method based on blocking and retrieval, and the method comprises the steps: employing the combination of a plurality of document analysis frameworks, and carrying out the preprocessing of documents in a plurality of formats, which are uploaded by a user; extracting and analyzing the document content, segmenting the document content into knowledge units with complete semantics through a document partitioning algorithm, creating indexes for the knowledge units, and constructing a knowledge base; a user question is input and submitted to an agricultural large model constructed based on a large language model LLM, a mixed retrieval strategy is adopted for a to-be-retrieved document, candidate knowledge fragments related to the question are preliminarily recalled, preliminary recall results are secondarily screened and sorted in a resorting mode, and an optimal knowledge fragment is output; through a preset prompt word engineering technology, a complete Prompt is generated, a retrieval evaluation data set is automatically generated, and an evaluation index is generated according to a retrieval result. According to the invention, the function of automatically generating and evaluating the test set is realized.

The invention relates to the technical field of electronic document content review methods, particularly discloses an electronic document content review method, and aims to solve the problems that traditional manual review is low in efficiency and inconsistent in standard and implicit compliance risks are difficult to recognize. The method comprises the following steps of: performing unified analysis and OCR (Optical Character Recognition) conversion on a PDF, DOCX or TXT format document, extracting a full-amount text, and structurally dividing the full-amount text into semantic units such as a title, a term and an attachment; in combination with large-model deep semantic analysis and double-layer regular rule matching, hidden risks such as logic conflicts and fuzzy expressions and dominant violations such as terms, units and sensitive words are recognized respectively; multi-source detection results are fused, after deduplication, sorting is carried out according to risk levels, and a structured report containing problem positioning, basis and suggestion is generated; according to the method, high-precision, high-efficiency and interpretable automatic review is realized through a'large model + rule 'dual mechanism, and compliance consistency and processing efficiency are remarkably improved.

The invention belongs to but is not limited to the technical field of artificial intelligence, and particularly relates to a multi-strategy RAG process optimization method and system for large model questioning and answering, and the method comprises the steps: carrying out the classification processing of a document of a user, and selecting semantic blocks or customizing the sizes of the blocks; receiving a text input by a user; the retrieval is divided into two parts, one part is semantic vector retrieval, and the other part is keyword retrieval; the system further screens and filters candidate text paragraphs in combination with structured meta-information on the basis of vector retrieval; after the preliminary vector retrieval, a bge-ryanker-v2-m3 model is introduced to carry out secondary sorting on the results; the model firstly receives input Prompt and converts the Prompt into vector representation; the vector is processed through a multi-layer Transform network so as to extract and model semantic features in the text; a corresponding output result is generated in combination with specific task requirements defined by Prompt; and outputting an answer generated by the large model for the user question.

Techniques for implementing hybrid retrieval augmented generation for rich document queries using a large language model (“LLM”) are disclosed. In an example method, a computing system receives documents which are added to databases. The computing system receives a query, from which tokens are generated. The computing system determines a first ranking of the documents using a probabilistic ranking method and a second ranking of the documents in response to a vector database query using an embedded representation. The computing system determines a third ranking of the documents based on the first ranking and the second ranking. The computing system selects top documents from the third ranking. The computing system determines a knowledge graph based on the top documents. The computing system outputs the query, the top documents, and at least a portion of the knowledge graph to an LLM. The computing system receives and outputs a response.

The invention relates to the field of engineering cost and technical and economic review, in particular to an intelligent technical and economic review method and system based on parameter fuzzy matching. And an intelligent process of equipment identification, model matching, unit conversion and quota review is designed, so that full-process automation of technical and economic review is realized. The system solves the problem of equipment name expression difference by utilizing fuzzy matching and a synonym library, the most adaptive equipment model is accurately selected through a mutual exclusion and scoring sorting mechanism, the accuracy and efficiency of review are remarkably improved, the problems of manual missed review, wrong review and different standards are effectively avoided, and in the quota matching process, the review efficiency is greatly improved. By applying keyword and antonym matching logic, a quota adjustment condition is automatically judged, a human-material-machine cost adjustment coefficient is calculated, and conversion from'experience judgment 'to'rule driving' is realized.

The invention provides an operation data analysis method and system, computer equipment and a storage medium, and belongs to the field of data processing, and the method comprises the steps: obtaining operation data of a plurality of shops; analyzing the operation data of the plurality of shops based on a preset short board index, and determining a short board type of each shop; sorting the plurality of stores according to the types of the short boards to obtain a processing priority sequence; according to the processing priority sequence, sequentially retrieving a strategy corresponding to each shop short board type from a preset knowledge base through an RAG technology; and constructing an operation suggestion of each shop according to the short board type of each shop and the corresponding strategy. According to the method, optimal configuration of analysis resources is realized, the timeliness of operation data analysis is further guaranteed, the generated operation suggestions have individuation and pertinence, and the operation data analysis effect is further guaranteed.

The invention provides a credible manuscript review decision-making method, system and equipment and a readable storage medium. The method comprises the following steps: receiving a manuscript to be reviewed, analyzing a document structure of the manuscript to be reviewed, and identifying a paper type; based on the identified paper type, calling a data processing agent, extracting a performance index corresponding to the paper type from at least one of a text, a table and a chart of the to-be-reviewed document, and converting the performance index into standardized structure data; traversing a preset database, comparing the data in the reference database with the structured performance indexes, and determining the similarity and the relative sequence of the performance indexes of the to-be-reviewed manuscript and the data in the database; and based on the similarity and the ranking percentage, outputting a quantitative score of the to-be-reviewed manuscript and the confidence of the quantitative score. Compared with the prior art, the method has the advantages that the professionality, objectivity and credibility of review are remarkably improved; and interference caused by training data prejudice and text subjective description is effectively reduced.

The invention provides a large language model retrieval enhancement generation method based on adaptive rewriting selection, and is suitable for the field of natural language processing and information retrieval. According to the method, a pre-trained large language model is introduced to automatically generate diversified rewriting queries, and a self-supervised rewriting sequencer is combined to perform correlation evaluation and sequencing on candidate rewriting statements. Through a context multi-arm bandit selector, the optimal rewriting number is dynamically determined according to query semantics, a high-quality rewriting subset is selected in a self-adaptive mode, and the coverage degree and precision of information retrieval are effectively improved. And the rewriting-driven knowledge retrieval module utilizes a plurality of high-quality rewriting, integration and deduplication related knowledge blocks in parallel, and continuously optimizes a Bandit strategy based on a feedback signal to realize online self-learning. Different from a traditional RAG system depending on fixed parameters and static rewriting, the method can intelligently adjust the retrieval process for complex or variable queries, and the accuracy and practicability of a retrieval enhancement generation system in an open domain and a multi-hop reasoning scene are remarkably improved. According to the method, efficient and flexible technical support is provided for intelligent question answering and knowledge discovery in a complex environment, and the application effect and popularization value of the large language model are greatly enhanced.