4 results about "Knowledge object" patented technology

This disclosure provides a method, computing device, computer program product, and computer-readable storage medium for constructing a knowledge graph of medical literature. The method includes: extracting multiple granularity knowledge objects from the medical literature using a medical domain-specific cue word template-driven large language model, wherein the multiple granularity knowledge objects include entities, relationships between entities, and declarations of each entity and / or relationship; determining key excerpts from the medical literature for each knowledge object using a progressively narrowing matching algorithm based on preceding and following anchor points; and classifying the knowledge objects according to the key excerpts of each knowledge object and the context of the corresponding medical literature using a predefined evidence-based medicine hierarchy to generate a knowledge graph of the medical literature.

The application provides a kind of spent fuel reprocessing process knowledge object processing method, device and equipment, wherein the method comprises: obtaining at least two knowledge objects in spent fuel reprocessing process;The knowledge object includes: at least one spent fuel reprocessing process, at least one equipment corresponding to the spent fuel reprocessing process, at least one process parameter corresponding to at least one equipment of the spent fuel reprocessing process;According to at least two knowledge objects, obtain the knowledge graph chromosome population of spent fuel reprocessing process;According to the preset algorithm, obtain the target chromosome individual vector of the knowledge graph chromosome population;According to the target chromosome individual vector, adjust the preset knowledge base of spent fuel reprocessing process, and obtain the target knowledge base.The application scheme improves the optimization ability of reasoning and problem solving of knowledge object in spent fuel reprocessing process, and improves the efficiency of spent fuel reprocessing process.

The application discloses a traditional Chinese medicine meridian and acupoint immersive teaching method and system based on three-dimensional modeling, relates to the technical field of three-dimensional modeling, and comprises the following steps: in response to a teaching display instruction, setting a lightweight neural network model on a teaching terminal, analyzing a 3D human body point cloud to be rendered, and constructing a meridian topology network; identifying meridian and acupoint knowledge objects based on the meridian topology network, combining a preset parameterized human body model, monitoring changes in bone joint poses and muscle envelope shapes, performing grid offset, generating a target skin area, and triggering a teaching interaction instruction; in response to the teaching interaction instruction, acquiring virtual interaction data, introducing a preset teaching rule engine, acquiring the positioning accuracy level of the meridian and acupoint and acupuncture actions, performing correction determination in combination with an abnormal operation event, and acquiring corrected three-dimensional modeling information; and the application improves the accuracy of acupoint positioning and the satisfaction of immersive teaching.

The application provides a petrochemical industry large model retrieval enhancement processing method and system. The method comprises the following steps: performing semantic unit segmentation, graphic element analysis, working condition slicing and object identification alignment on various data to generate a standardized industrial knowledge object set; constructing an industry knowledge base associated with process topology domain, control logic domain, risk event domain and system version domain based on the standardized industrial knowledge object set; receiving a retrieval enhancement request, analyzing to obtain object entities, behavior intentions, time constraints and safety levels, and generating a query semantic skeleton; performing error correction re-retrieval according to the retrieval quality evaluation result to generate a candidate evidence cluster; determining a target evidence cluster, generating a candidate response based on the target evidence cluster and a preset constraint, and performing interlocking conflict verification, process boundary verification and system applicability verification on the candidate response, and outputting a retrieval enhancement processing result. The application can improve retrieval accuracy, enhance working condition version adaptability and improve output safety.