13 results about "Text database" patented technology

The application discloses a kind of civil aviation safety report theme modeling methods based on large language model and semantic enhancement, its method includes: civil aviation safety theme semantic feature model extracts the text semantic vector and event structure vector of safety report text, and obtains the fusion feature vector of safety report text by fusion;Safety report text database is carried out cluster clustering processing and obtains initial cluster set and noise sample set;Select representative front p% as the representative sample set of cluster;Construct noise sample evaluation repair mechanism module, and select the candidate cluster to which noise sample belongs using noise sample evaluation repair mechanism module;Comprehensive gain function is constructed using representative sample set in cluster, and representative sample iteration screening processing of candidate sample is carried out in cluster.The application realizes the theme modeling goal of semantic accuracy, comprehensive coverage and stable result by multi-module collaborative innovation, and provides reliable technical support for management decision.

This invention discloses an entity relationship recognition method and apparatus for decision knowledge graphs. The method includes: retrieving a set of decision text information from a decision knowledge text database; preprocessing the set of decision text information to obtain a preprocessed dataset; and performing entity relationship recognition processing on the preprocessed dataset to obtain an entity relationship information set.

This invention discloses a method and system for generating auxiliary diagnostic results and reference prescriptions. First, a medical record database and a treatment strategy database are constructed and stored in a vector database and a text retrieval database, respectively. For a target medical record, vectorization and textualization are performed. Then, candidate medical records and treatment strategies are retrieved from the medical record vector database, medical record text database, treatment strategy vector database, and treatment strategy text database via four channels. Subsequently, the RRF algorithm is used to fuse and sort the vector and text channels, and the candidate results are further refined based on a pre-trained semantic ranking model to obtain a small number of reference medical records and treatment strategies highly relevant to the current condition. Finally, prompt words are constructed by combining the target medical record, reference medical records, and treatment strategies to generate candidate results such as TCM diagnosis, TCM syndrome type, Western medicine diagnosis, and TCM prescriptions, serving as auxiliary references for doctors' clinical decision-making. This significantly improves the accuracy, interpretability, and efficiency of intelligent assisted diagnosis and treatment.

This invention discloses a method for mitigating hallucinations in large-scale narratives based on multi-agent collaboration and dynamic state tracking. The method includes: generating an event graph composed of multiple atomic events using an outline agent; mapping these atomic events from an event occurrence time series to a text narrative time series using an orchestration agent; constructing a non-linear narrative outline; retrieving historical fragments related to the semantics and sentiment of the current chapter from the generated historical text; generating the text content of the current chapter using a writing agent; performing real-time logical consistency checks on the content during generation; triggering a correction mechanism when entity state changes in the generated content violate preset state transition rules; and parsing the entity state changes after the current chapter is finalized, updating the global state table and historical text database. This invention effectively mitigates hallucinations in long narratives, significantly improving the logical rigor and long-term coherence of the plot development while maintaining the literary quality of the text.

This invention discloses an event-driven incremental knowledge extraction and fusion method and apparatus. The method includes: retrieving a set of factual text information from a text database; preprocessing the set of factual text information to obtain a set of factual text information; and performing knowledge fusion processing on the set of factual text information to obtain a set of factual knowledge ontology information.

This invention provides a method, system, device, and medium for compliance risk avoidance in generative artificial intelligence. The method includes: acquiring user input text; preprocessing the input text and calculating its compliance potential value, wherein the compliance potential value is positively correlated with the compliance relevance, discourse empowerment, and group influence of each character in the input text; based on the calculated compliance potential value, using a normalization algorithm to calculate an empirical threshold for the input text, and dividing the input text into three levels according to the empirical threshold; constructing a three-level text database; generating a final question-and-answer result by calling the three-level text database according to the level of the empirical threshold corresponding to the input text; and displaying the final question-and-answer result to the user. This addresses how to ensure that the generated content does not have content compliance issues while providing users with more authoritative, comprehensive, and reliable answers when applying generative artificial intelligence in areas involving compliance expression related to content compliance.

The application relates to a data processing system for updating a word library, which comprises a preset word library, a processor and a memory storing a computer program, and when the computer program is executed by the processor, the following steps are realized: a candidate text set is acquired, an intermediate text set is acquired, a specific word list corresponding to the candidate text set is acquired according to the candidate text set and the intermediate text set, a target text set corresponding to the specific word is acquired, a target label list is acquired, a candidate label set is acquired according to the target text set and the target label list, a key label set corresponding to the specific word list is acquired, each key label list in the key label set is subjected to a deduplication treatment, and a label corresponding to each specific word in the specific word list is acquired to update the preset word library. The application is not limited to a same text database in terms of a text source, can be compared with other text databases, the accuracy of the acquired specific word is improved, and the accuracy of the acquired updated word library is high.

This disclosure relates to a communication server apparatus, a communication device, and a method of operating the same. A communication server apparatus (100) is configured to receive (202) text data comprising at least one text data element associated with an abbreviated text unit. The text data element is compared (204) with a plurality of candidate text data elements from a given text database, each candidate text data element being associated with a corresponding candidate text unit in the database. A similarity metric between the at least one text data element and these candidate text data elements is determined (206), and the candidate text data elements are processed (208) to select candidate text data elements that have an ordered relationship with the abbreviated text unit. These similarity metric values ​​and the selection of these candidate text data elements are used (210) to designate the associated candidate text unit as the de-abbreviated text unit of the abbreviated text unit.

[Problem] To provide a large language model using retrieval-augmented generation (RAG) that achieves: scaling up a capacity for recording additional information by constructing a plurality of RAG databases respectively for relevant fields as necessary; eliminating missed references by referring to the entire text or image of an original document page that includes a description of relevant matter on the basis of chunks obtained by RAG retrieval, and at the same time, reflecting, in the context of a prompt, background information or relevant information that may be described around a retrieved chunk; and eliminating the need for costly and time-consuming re-response generation by putting records of prompts and responses into a database to make the records retrievable. [Solution] A RAG database of the invention includes a page image acquisition means, a page text database recording means, a relevant feature vector extraction means, and a feature vector database recording means.

This invention discloses a method for constructing a domain-specific large language model fine-tuning dataset based on multi-database linkage, comprising the following steps: collecting structured business data, unstructured text data, and knowledge graph data to form a structured business database, an unstructured text database, and a knowledge graph database, respectively; establishing a multi-database linkage and association mechanism; establishing a dynamic priority mechanism to resolve conflicts according to priority when different data sources conflict; generating structured instruction question-and-answer samples using quadruples based on the multi-database linkage and association results; and using a large language model for diversified rewriting and quality filtering to construct a domain-specific large language model fine-tuning dataset. This invention effectively solves the technical problem of inconsistencies between historical data and current business rules by using data priority for conflict resolution; and significantly improves sample quality and generation efficiency by using quadruples to generate structured instruction question-and-answer samples.

This application provides a retrieval enhancement generation method, apparatus, device, medium, and program product, relating to the field of artificial intelligence technology, for improving the quality of generated answers. The specific technical solution is as follows: A user-inputted query question is input into an answer model; the answer model obtains modal requirement information corresponding to the user-inputted query question; the answer model performs a text query in a text database based on the query question to obtain at least one response text fragment; the answer model performs contextual reconstruction processing on the at least one response text fragment based on a target reconstruction strategy matching the modal requirement information to obtain a target answer that conforms to the user's modal requirement information. This application is applied to intelligent search scenarios.

The application discloses a retrieval enhancement method and system based on asymmetric local sensitive hashing. The retrieval enhancement method comprises the following steps: establishing a knowledge text database, including a plurality of knowledge texts; acquiring a query text, and extracting target knowledge texts related to the query text in the knowledge text database based on an asymmetric local sensitive hashing algorithm; establishing a background formula, the background formula including placeholders uniquely corresponding to the target knowledge texts and the query text, and then replacing the corresponding placeholders in the background formula with the target knowledge texts and the query text to obtain an input text; and inputting the input text into a large language model to output an answer text. The application has the characteristics of accuracy and efficiency.