533 results about "Named-entity recognition" patented technology

The invention belongs to the technical field of natural language processing and multi-modal information extraction, and particularly relates to a multi-modal named entity recognition method based on semantic alignment and cross-modal graph fusion, which comprises the following steps: S1, acquiring a data sample containing a text sequence and image content; s2, encoding the text and the image into vectors respectively; s3, similarity is calculated through a trainable bilinear function, and optimization is carried out through loss comparison; s4, cross-modal attention is used to enhance association information between modals; s5, determining the proportion of reserved image information through a modal matching module; s6, introducing a gating mechanism to dynamically fuse visual and text features; s7, realizing local and global information complementation by a cross-modal graph fusion model; and S8, inputting the fused representation into the CRF layer to predict the entity type. According to the method, fine semantic alignment can be realized in a weak image-text correlation context, and balance between local entity recognition and global semantic understanding can be achieved.

The invention discloses an AI-based large-model-driven contract review and regulation interpretation method and system, and the method comprises the steps: S1, collecting a regulation and institute document, and carrying out the text extraction and semantic disassembly, and obtaining term information; s2, based on a LawCheckLM + BERT-CRF named entity recognition model, performing entity recognition and classification labeling on clause information, and storing recognized key information fields into a knowledge base; s3, encoding each piece of clause information into a semantic vector through an embedded model, and storing the semantic vector into a knowledge base; and S4, performing text extraction and semantic disassembly on the uploaded contract document to obtain clause information, repeating the steps S2-S3, retrieving similar semantic vectors of laws and regulations or institutional documents similar to the contract from the knowledge base as references, if the similar semantic vectors are not retrieved, taking Top-N laws and regulations or institutional documents with similar semantics as outputs, and marking the Top-N laws and regulations or institutional documents as required to be examined. The problems that an existing management system is difficult to adapt to regulation changes, complex contract review and cross-scene deployment are achieved, and the overall iteration period is long are solved.

Artificially intelligent systems and methods for financial coaching provide personalized, fiduciary-compliant financial guidance through advanced machine learning architectures with measurable performance criteria. The systems implement privacy-preserving processing pipelines that detect personally identifiable information using multi-layered pattern recognition including regular expressions for formatted data sequences, named entity recognition with confidence thresholds above 0.85, and contextual analysis algorithms. A multi-step artificial intelligence processing workflow includes automated language detection, emotional tone classification with confidence scoring, financial profile transformation using predefined templates, context-aware question rephrasing, and semantic similarity matching employing vector embeddings with financial domain vocabulary weighting applying multiplier values between 1.3-2.0. Specialized training methodologies expand datasets through mathematical transformation functions utilizing statistical standard deviations with incremental variations between 0.5-2.0. Mood-based escalation logic automatically transfers users to human advisors when emotional indicators exceed confidence thresholds above 0.8. The systems maintain response times below 5 seconds while providing regulatory compliance through curated content sources and predefined fiduciary instruction parameters.

The invention relates to a small-sample contrast enhancement fine tuning method and system based on a large language model, which are used for identifying named entities in recruitment texts. The method comprises the steps of performing cleaning and format conversion on an original recruitment text, and generating an input sample conforming to a natural language instruction format; under the condition that the labeled samples are insufficient, positive and negative sample pairs are constructed to enhance the recognition capability of the model on entity categories and boundaries; carrying out low-rank parameter updating on the pre-trained large language model by adopting a LoRA fine tuning technology, and reducing computing resource consumption in combination with 4-bit quantitative training; in a pre-training large language model reasoning process, through a multi-dimensional joint confidence evaluation mechanism, confidence of four dimensions of entity levels, lengths, types and contexts is synthesized, and low-confidence identification results are filtered after dynamic weighted normalization processing. The method is suitable for recruitment recommendation, talent matching and other downstream tasks, and has the advantages of high recognition accuracy, low training cost, high system robustness and the like.

The invention relates to the technical field of information extraction, in particular to an information extraction method and system for electronic archives. The method comprises the following steps: obtaining a to-be-processed electronic file and carrying out OCR identification to generate initial text data; error detection is carried out on the initial text data, and OCR error recognition candidate items in the initial text data are recognized; for each OCR misrecognition candidate item, generating a first data name according to context semantics and a layout structure of the candidate item; extracting low-level features of the first data name, and performing named entity recognition on each OCR misrecognition candidate item by utilizing a preset field word list in combination with the first data name; through a four-in-one process of ''misrecognition detection + named entity recognition + semantic error correction + templated extraction'', the core technology bottlenecks of inaccurate recognition, poor error correction capability, low information extraction intelligence and the like in the prior art are solved, and the accuracy, stability and intelligent level of electronic archive information extraction are remarkably improved.

The invention relates to the field of artificial intelligence, in particular to a long text segmentation information retrieval result coherence enhancement method, system and equipment based on a knowledge graph, and the method comprises the following steps: receiving a long text input by a user; segmenting the long text into a plurality of segments according to a preset length, and recording the original position and sequence of each segment; performing named entity recognition on each fragment, and extracting a key entity; querying related information in a knowledge graph according to the key entity, and obtaining attributes of the entity and a relationship between the attributes; integrating the related information obtained by query into the corresponding text fragment to form a context enhanced fragment; receiving a user query request; calculating a correlation score based on the context enhanced fragment and the user query, and retrieving a related text fragment by using a hybrid retrieval strategy; and selecting the related text fragment with the highest score to generate a final answer, and outputting the final answer to the user. Therefore, the information integrity and the context continuity in the retrieval process are ensured.

The invention relates to the technical field of data processing, in particular to an integrated multi-mode culture resource intelligent data governance and management system. The system comprises a data acquisition module, an intelligent processing module, a data management module, a knowledge organization module and an interactive display module. Synchronous acquisition of videos, audios, images and three-dimensional point clouds is realized through camera equipment, audio acquisition equipment and a laser scanner, and cultural elements are extracted through action recognition, speech recognition model processing dialect transcription, image semantic segmentation and named entity recognition by adopting a convolutional neural network. The system realizes structured description and semantic mapping of data on the basis of non-abandoned domain ontology, and constructs a knowledge graph by means of a graph database and a graph neural network. And finally, realizing three-dimensional visual display of the cultural resources by combining a virtual reality technology. According to the method, the problems of non-abandoned multi-modal data processing splitting, insufficient semantic organization and weak display interactivity in the prior art are solved, and the digital governance level and the propagation capability of cultural resources are improved.

The invention provides a method and equipment for constructing a text entity recognition model based on large model data enhancement. The method comprises the following steps: firstly, constructing an initial model comprising a preprocessing unit, a syntax dependency analysis unit, a data correction enhancement unit, a context coding unit, a syntax enhancement unit, an expression fusion unit and a sequence decoding processing unit; preprocessing the training sample to obtain a preprocessed text, and establishing a preliminary dependency graph through syntactic analysis; correcting and enhancing the text and the dependency graph by the large language model to obtain a text sequence and a dependency graph for subsequent use; encoding the text sequence to obtain an initial representation vector, and enhancing the vector in combination with the dependency graph; after fusion, decoding and outputting a prediction label containing a lexical entity label; and calculating loss by using a function containing conditional random field structure loss, judging convergence, and if not, updating parameters and continuing training until a target model is obtained. According to the method, data are optimized and expanded by means of a large language model, syntactic dependency enhancement features are combined, and the recognition capability of the complex text named entities is improved.

The invention relates to a forest fire knowledge modeling method based on named entity recognition and relation extraction, and the method specifically comprises the following steps: firstly collecting original text data in the field of forest fire, and constructing a text data set through preprocessing and entity and relation labeling; constructing a named entity recognition and relation extraction model, and realizing entity recognition and relation extraction of forest fire knowledge through multi-granularity adaptive word segmentation, dynamic standardization and pre-training language model encoder feature enhancement to obtain triple global confidence; adopting a multi-task joint training framework to calculate a loss function of the model, and performing end-to-end model parameter optimization on the model based on the loss function; inputting a new forest fire text into the trained model, and screening a high-confidence relation triple; and finally, constructing the structured forest fire field knowledge graph based on the extracted triple. According to the method, forest fire field knowledge can be efficiently excavated, and accurate knowledge support is provided for forest fire prevention and control and the like.

The invention relates to a diabetes mellitus question-answering method, system and device based on a knowledge graph and a medium, and belongs to the technical field of medical intelligent question-answering. The diabetes mellitus question-answering method based on the knowledge graph analyzes a question sentence of a user based on a pre-trained intention classification model so as to identify a user intention; analyzing the user question based on a pre-trained named entity recognition model to recognize a question entity; performing dynamic query on the constructed diabetes knowledge graph based on the user intention and the question entity to obtain a knowledge graph query result; encoding the user question by adopting a retrieval enhancement generation model, and performing semantic retrieval on the constructed diabetes semantic retrieval library based on the encoded user question to obtain related knowledge corresponding to the user question; and the knowledge graph query result and related knowledge are input into the constructed large language model to obtain the diabetes mellitus question and answer, so that the accuracy and specialty of the answer are improved.

The invention relates to the technical field of bidding and tendering information processing, in particular to a bidding and tendering template generation method based on a large language model, which comprises the following steps: receiving and analyzing bidding and tendering project demand information input by a user, extracting key elements through a named entity recognition model, and generating standardized demand data; inputting the standardized demand data into a pre-trained large language model to generate an initial bidding and tendering template text; based on a bidding and tendering clause logic rule base, internal logic consistency scanning is carried out on the initial template text, and potential conflict clauses are identified and marked; and generating an optimization instruction according to a conflict detection result, driving the large language model to complete directional revision of conflict terms, and finally outputting a target bidding and tendering template with a standard structure and consistent logic. The method has the capabilities of efficient generation, logic self-inspection and automatic repair, and the intelligent level and compliance control capability of bidding document compilation are remarkably improved.

The invention relates to the technical field of information security, in particular to a user information protection method and device based on a dynamic desensitization strategy, and the method comprises the steps: determining sensitive data and a sensitivity score thereof through a named entity recognition model and a sensitivity scoring algorithm, extracting scene parameters, and determining a scene risk score through a risk quantification algorithm; and then a dynamic desensitization strategy generation algorithm is called to generate a target desensitization strategy, and finally, the strategy is used for desensitization and a reversible desensitization data identifier is generated. Through the series of steps, the desensitization strategy is dynamically generated according to different use scenes, the problem of scene adaptability deficiency caused by static desensitization strategy stiffness is solved, meanwhile, the reversible desensitization reduces the security risk caused by reverse cracking of rules, excessive desensitization is avoided, and the data availability is improved.

The invention provides a paper file information input method, system and device and a medium, and belongs to the technical field of file information processing. The method comprises the steps of collecting a paper file image, performing preprocessing in sequence to obtain a preprocessed image, and performing region segmentation on the preprocessed image by using a target detection algorithm to generate a region coordinate mapping table; generating a file template by using a predefined file template library according to the region coordinate mapping table; performing printing form recognition and handwritten form recognition on characters in the image by using a sequence generation model to generate a text sequence, and recognizing a table structure by using a two-dimensional convolutional network; according to the text sequence and the table structure data, a named entity recognition model is used for labeling field types, data formats are checked and corrected according to template rules, errors are corrected in combination with context semantic similarity, and structured data are generated; and mapping the structured data to an information system database table according to a template field mapping rule, and inputting the structured data into an information system.

The invention discloses a medical review and publication quality control system based on deep learning and medical knowledge, and relates to the technical field of medical information processing, and the system comprises a medical knowledge enhancement module which is used for carrying out the named entity recognition and relation extraction technology based on a multi-source fusion medical knowledge system, constructing a structured medical knowledge graph and carrying out intensive training; the multi-dimensional examination and analysis module is used for performing multi-dimensional evaluation on the manuscript based on the medical knowledge graph; the intelligent interaction and feedback module is used for displaying the evaluation result of each dimension through a visual interface; and the publishing quality management and control module is used for sequentially executing whole-process publishing supervision of standardized monitoring, quality inspection and abnormal intervention on the manuscripts passing the examination and analysis. The manuscript reviewing efficiency is greatly improved, and the requirements of high-frequency and rapid publishing of medical scientific research are met; the problem of review difference caused by subjective factors of a manuscript reviewer is solved, and the stability and reliability of the assessment result of the same manuscript are ensured through a standardized and quantitative assessment system.

The invention provides a metro industry knowledge file reading method, which belongs to the technical field of file reading, and comprises the following steps: processing input metro industry knowledge files in various formats, and outputting uniformly formatted intermediate text data; performing word segmentation and part-of-speech tagging, domain term recognition and enhancement, named entity recognition, relation extraction and key information extraction on the intermediate text data; constructing a metro field knowledge graph based on the obtained entities, relationships and key information, storing the structured key information into a structured database, and establishing a graph node and index link to obtain a knowledge base fusing structured knowledge and unstructured document indexes; and performing query understanding on the user query and performing retrieval based on the knowledge base. The query accuracy and knowledge relevance are improved, and the efficiency of obtaining knowledge by subway workers is remarkably improved.

The invention provides an electric power industry-oriented A I intelligent question answering and professional report generation method, which comprises the steps of extracting key information from a constructed knowledge graph to form a preliminary electric power knowledge base structure, and the extraction process comprises the steps of identifying high-frequency nodes, analyzing connectivity among the nodes and evaluating importance weights of the nodes; based on the preliminary knowledge base structure, integrating structured and unstructured data in the power field to form a complete power knowledge base, and keeping synchronous updating with the knowledge graph; based on the electric power knowledge base, training a natural language processing model in the electric power field, which comprises named entity recognition and relation extraction by adopting a deep learning model in sequence; and performing semantic annotation on nodes and edges in the knowledge graph by utilizing a trained natural language processing model, and integrating new semantic information into the knowledge graph through entity alignment and relation mapping.

The invention discloses a retrieval method based on a semantic enhanced knowledge graph, which relates to the technical field of information, and comprises the following steps: receiving a natural language query of a user, and carrying out deep analysis on the query, including named entity recognition and linking, relationship extraction and query intention classification; and based on an analysis result, extracting a related local sub-graph from the knowledge graph, generating a query context vector, and generating dynamic semantic embedding for the sub-graph through a query-perceived graph attention network to obtain a dynamic enhanced semantic graph. According to the retrieval method based on the semantic enhancement knowledge graph, the retrieval precision and the recall rate are remarkably improved, the limitation of static knowledge representation is solved through a dynamic semantic enhancement mechanism of query intention perception, so that the local semantic representation of the knowledge graph is highly aligned with the query intention of a specific user; and the ability of understanding and answering complex, fuzzy, ambiguous and multi-hop queries is improved.

The invention provides a planning analysis workflow intelligent arrangement method and device based on a natural language instruction, and the method and device can deeply understand the complex analysis intention of a plan in a natural language through a mixed analysis engine comprising a large-scale language model and a named entity recognition model. And according to the understood intention, a proper operator and a proper model are autonomously selected from a tool library in combination with a planning analysis knowledge graph, so that a complete, logic self-consistent and executable multi-step analysis workflow can be autonomously and dynamically arranged, the workflow is automatically executed, and multi-modal data is collected. According to the method and the device, the technical problems in multiple aspects from keyword matching to composite intention deep understanding, from artificial process design to knowledge-driven intelligent arrangement and from one-way instruction execution to interactive exploration analysis in the prior art are solved, and the method and the device have the advantages that the technical problems in multiple aspects from keyword matching to composite intention deep understanding, from artificial process design to knowledge-driven intelligent arrangement and from one-way instruction execution to interactive exploration analysis are solved; the planning work efficiency can be improved, the technical threshold is reduced, and industry digital transformation is promoted.

The invention belongs to the technical field of artificial intelligence, and discloses a CNN-based call quality inspection method, device and equipment and a storage medium, the CNN-based call quality inspection method comprises the following steps: inputting a word vector sequence obtained by converting a target text corresponding to a target seat call into a preset sentiment analysis model to obtain a sentiment tag corresponding to the target text; analyzing the importance degree of each word vector in the word vector sequence to obtain a keyword vector in the word vector sequence; performing named entity recognition on each word vector in the word vector sequence to obtain an optimal entity tag sequence corresponding to the word vector sequence; and splicing the audio feature, the emotion tag, the keyword vector and the optimal entity tag sequence, inputting the obtained multi-modal fusion feature into a preset evaluation model, and outputting to obtain a quality inspection result. The method and the system can be applied to business management systems of financial science and technology, medical health and the like, and solve the technical problem that efficiency and quality cannot be considered in a call quality inspection mode based on the prior art.

The invention discloses a structured information extraction system for a mental disorder electronic medical record based on a knowledge graph and a natural language, and belongs to the technical field of medical information processing. According to the system, firstly, core concepts such as diseases, symptoms and drugs are extracted from an authoritative guide to construct a mental disorder knowledge graph, and a standardized clinical knowledge base is established; then pre-training and fine-tuning the deep learning model on a large number of biomedical texts and desensitized medical records to enable the deep learning model to have a medical language understanding ability; performing preprocessing, named entity recognition and entity linking on the electronic medical record text, and mapping spoken expressions to standard medical terms; inference is carried out by utilizing a relation extraction model and combining with a knowledge graph to complement implicit clinical information; and finally, structured data output meeting the standards of FHIR and the like is generated. According to the method, through deep fusion of knowledge driving and data driving, the problems of insufficient semantic understanding and weak generalization ability of a traditional method are effectively solved, and the accuracy and clinical value of electronic medical record structured information extraction are remarkably improved.

The invention discloses a structured data storage method based on natural language transformation, which comprises the following steps of: a system initialization configuration stage: deploying a protocol adapter in a local memory of a PC (Personal Computer) client, and loading natural language processing pipeline configuration parameters; a heterogeneous data acquisition stage: capturing a multi-source text data stream through the protocol adapter, uniformly converting the multi-source text data stream into a standardized data packet, and sending the standardized data packet to a message queue theme; a text cleaning stage: a named entity recognition stage: inputting the pure text data into an NER module deployed with a language model loader; in the conditional feature extraction stage, feature vectors are generated for texts meeting preset conditions on the basis of entity type tags in the entity recognition result; and a consistent storage stage: inserting the entity identification results into a relational database in batches, and updating the entity mapping relationship in the cache. According to the method, the intelligent level of cache management is remarkably improved, and the access fluency of the key data of the user is guaranteed.

The invention discloses an intelligent system and method for realizing multi-mode psychological counseling and seamless connection based on an education robot. A multi-mode psychological counseling system is constructed aiming at the problems of insufficient privacy protection of a traditional campus psychological counseling room, shortage of psychological counseling resources of mountainous schools and the like. The system provides psychological counseling modes such as an education robot, real person remote and man-machine collaboration, can be combined with counseling environments such as a simple psychological counseling room and a VR counseling room to become multi-modal psychological counseling, supports seamless switching of each modal and spatial form, and achieves the intelligent scheduling technology through a multi-modal data real-time synchronization engine and reinforcement learning driving. And data encryption transmission and millisecond-level service form switching are realized. A dynamic auxiliary screen system is innovatively integrated, and psychological archives are dynamically associated with a knowledge graph through a domain-customized named entity recognition model (NER, the accuracy rate is greater than or equal to 95%), so that the remote real expert key information retrieval time is less than or equal to 2 seconds, and the decision-making efficiency is improved by 78%. The system meets the requirements of various laws and regulations such as the personal information protection law.

The invention belongs to the technical field of natural language processing and power system intelligent information processing. Comprising the following steps: S1, preparing data; s2, performing named entity recognition and relation extraction on the text block by utilizing a pre-trained large language model; s3, constructing a knowledge graph in the urban power distribution field by using the triple summary; s4, generating a semantic abstract with a hierarchical structure for each theme community; s5, querying and retrieving: retrieving related information from the vector index and the knowledge graph in parallel by receiving a natural language query of a user, and fusing results into a context knowledge set; and S6, an answer generation step: inputting the query and the context knowledge set thereof into a large language model, and generating an answer which is coherent in semantics, accurate in facts and covers query requirements through a preset prompt strategy guide model. According to the method, the accuracy and practicability of the question-answering system are remarkably improved, and core technical support is provided for intelligent operation and maintenance of the power distribution network.

The invention discloses a natural language processing technology-based cultural and creative element data analysis method, which comprises the following steps of: based on AI intelligent plush doll interaction data, generating a cultural and creative structure embedding degree index through named entity recognition and semantic network topology analysis; an impulse response function is constructed based on the cultural creative structure embedding degree, and a cultural creative memory cycle index is calculated in combination with the path length and information backflow; fusing the multi-modal features and the ideal innovation region, and outputting a perception novelty score; screening high-potential-energy elements to construct a semantic sub-graph, simulating a thermodynamic process, and identifying a text-creative fusion hotspot chain; and decoding the hotspot chain into a design semantic tag, matching a material database to generate an appearance scheme, and converting the appearance scheme into an interaction template, thereby realizing a customized design closed loop from a user dialogue to a doll form and behavior. Through data driving and cross-modal analysis, implicit culture preferences in user dialogues are dominated, and creativity of the appearance of the AI intelligent plush doll and humanization of interaction are improved.

The invention discloses a flexible business process automatic configuration method and system based on semantic intention recognition, and relates to the technical field of semantic recognition, and the method comprises the steps of semantic fragment extraction, process fragment atlas construction, process structure generation, process configuration visualization and flexible business process configuration. The method comprises the following steps: carrying out word segmentation, named entity recognition, semantic role labeling and relation extraction on natural language input by adopting a hierarchical guide type semantic decomposition combined extraction model, establishing a business semantic labeling system, and realizing modeling and extraction of semantic elements such as actions, objects, roles and conditions; on the basis, a process fragment graph improved by structural semantic fusion is constructed, and through node-level splitting, semantic tag binding and logic connection abstraction, process fragment graph data with tag attributes are generated, so that semantic understanding, process matching and cross-scene multiplexing capabilities are improved, and efficient, flexible and intelligent configuration of a business process is realized.

The invention relates to a medical bill intelligent processing method, system and device and a medium. The method comprises the following steps: performing type identification and text structuring processing on a medical bill image through an image classification model and an OCR technology to obtain a preliminary identification result; key entities such as diseases, drugs and inspection items are extracted through a medical field named entity recognition model, entity standardization is achieved through a standard medical term library, cross-bill information of the same patient is associated, and a medical entity set is generated; constructing a doctor seeing event knowledge graph in combination with preset medical rules, wherein nodes of the doctor seeing event knowledge graph comprise standardized term attributes, and timestamp weights are embedded into relation edges; and finally, based on a timestamp weight and a graph traversal algorithm, detecting three types of core conflicts including drug taboo, diagnosis and treatment contradictions and time sequence anomalies, calculating risk levels in combination with influence factors, outputting a structured conflict report, and realizing deep semantic association and cross-document logic conflict detection of the multi-source medical bills. And the intelligent level and the risk identification capability of medical bill processing are improved.

The invention relates to the technical field of natural language processing, in particular to a data enhancement and feature enhancement-based Chinese named entity recognition method, which is based on a feature enhancement double-attention named entity recognition model, and is characterized in that the model comprises an embedded layer, a data enhancement module, a multi-scale convolution fusion attention layer and a prediction layer; the embedded layer uses a dual-channel attention fusion module to process texts in parallel, fuses multi-dimensional information features of Chinese characters, combines an error correction type mask language model, a pre-training model and a bidirectional gating loop unit, fuses local and global text features, and obtains text representation from multiple dimensions and multiple levels; according to the method, by introducing an innovative model mechanism or training strategy, challenges such as label sparsity and text noise existing in a Chinese named entity recognition task can be effectively handled, the learning ability and generalization performance of the model for long-tail entity categories are improved, and the robustness of the model in a real and non-ideal data environment is enhanced.

The invention discloses an equipment fault diagnosis method and device based on a dynamic knowledge graph and a large model fine tuning technology. The method comprises the following steps: firstly, identifying a core entity from multi-source heterogeneous equipment fault data through a named entity identification model for fine tuning of domain data and a relation extraction model for special fine tuning of a fault diagnosis domain corpus, mining deep semantic association, and injecting the deep semantic association into a graph database after cleaning to form an initial knowledge graph; receiving user natural language fault description, realizing term and standard entity linking through editing distance fuzzy matching and Sension-BERT semantic vector similarity calculation, and combining bidirectional retrieval and attention mechanism fusion to obtain an enhanced context; and finally, generating a structured diagnosis report containing thinking chain reasoning based on an enhanced context by utilizing a specialized fine-tuning fault diagnosis large language model. According to the method, the limitation of a traditional diagnosis method is effectively solved, high-precision and interpretable equipment fault diagnosis is realized, and the diagnosis efficiency and reliability are improved.

The invention belongs to the technical field of natural language processing and artificial intelligence, and relates to a biomedicine named entity recognition method based on causal graph guided anti-fact analysis, which comprises the specific steps of causal graph construction, false node anti-fact analysis, false link anti-fact analysis, consistency constraint and model training and optimization. According to the method, a dual anti-fact analysis mechanism based on syntactic analysis and adversarial disturbance is introduced into large-scale language model training, so that false factors in input features can be effectively identified and intervened. Therefore, the model can focus on core features with real causal association in an entity identification task, so that the robustness and the cross-domain generalization ability of the model are remarkably enhanced. Experimental results prove that the Micro-F1 scores on a plurality of biomedical named entity recognition reference data sets are all advanced to the prior art.