13 results about "Word sense" patented technology

The application discloses a sense vector generation and language model intermediate layer enhancement method and device, relates to the technical field of natural language processing and semantic representation learning. The method comprises the following steps: obtaining an unlabeled text corpus and performing preprocessing to obtain a word sequence; performing sense subspace division on a word vector corresponding to the word sequence, constructing a sense matching matrix between a central word sense subvector and a context word sense subvector, and determining a maximum active sense pair in a training process; meanwhile, negative sample maximum matching inhibition, non-active sense combination constraint and sense subvector orthogonal constraint are combined to update a central word embedding matrix, so that a sense vector library is constructed; and based on the sense vector library, a downstream task is processed through a language model intermediate layer sense semantic enhancement method to generate a downstream task result. The embodiment of the application can improve the accuracy and distinguishability of polysemous word sense representation.

The present invention discloses a public opinion detection and analysis system thereof, which relates to the technical field of public opinion detection and analysis, and includes: a word sense association module, which is used to construct a word sense association tree of characters, wherein the root node is any character, the child nodes are characters related to the root node, and the child nodes are assigned values ​​according to the co-occurrence frequency, so that the sum of the values ​​of all child nodes is one; a semantic association module, which is used to construct a word sense association tree of each child node based on the child node of the word sense association tree as the root node, until a preset number of layers is reached to obtain a semantic association tree; a keyword detection module, which is used to match and analyze the semantic association trees of the document to be analyzed and the keyword, and obtain a set of public opinion keywords for the document to be analyzed. By constructing a word association tree, a co-occurrence frequency matrix and a multi-level similarity matching calculation model, high accuracy, dynamic adaptability and high robustness of keyword recognition are achieved, and the accuracy and stability of keyword recognition in complex contexts are improved.

The application provides a word sense disambiguation method and device, wherein the method comprises the following steps: determining a to-be-disambiguated entity existing in to-be-disambiguated text and a corresponding candidate entity list based on an RPA knowledge graph; determining embedding features corresponding to each candidate entity in the candidate entity list and the to-be-disambiguated entity through RPA feature extraction based on each candidate entity and the to-be-disambiguated entity; and determining whether the to-be-disambiguated entity and the candidate entity are the same entity based on a word sense disambiguation model, the embedding features corresponding to each candidate entity, and the embedding features corresponding to the to-be-disambiguated entity. The application realizes the comparison of embedding features of the to-be-disambiguated text and the candidate text by comprehensively embedding entity features, entity context features and word features, determines whether the to-be-disambiguated entity and the candidate entity are the same entity, and obtains more rich and comprehensive text information, which is conducive to accurately analyzing the word sense and improving the word sense disambiguation accuracy.

The invention relates to a character recognition method and system, and the method comprises the steps: inputting a target image comprising a to-be-recognized statement into a trained character recognition model, and obtaining an initial recognition result; the to-be-recognized statement comprises a plurality of to-be-recognized characters; the initial recognition result comprises a result character with the highest result probability corresponding to each character to be recognized; inputting the plurality of result characters into a trained context word meaning scoring model to obtain a word meaning score of each result character; and obtaining a target result character of each to-be-recognized character according to the word meaning score and the result probability. According to the method, the target result character of the to-be-recognized character is obtained by combining character recognition and context semantic scoring, and character recognition of the statement is performed by integrating the character recognition dimension and the semantic analysis dimension, so that the accuracy of character recognition can be improved.

Inventive name: A word sense disambiguation method based on multi-source knowledge fusion and dynamic enhancement Abstract: The application discloses a word sense disambiguation method based on multi-source knowledge fusion and dynamic enhancement, and belongs to the technical field of natural language processing. The method comprises the following steps: obtaining a text to be disambiguated and performing cleaning, word segmentation and annotation processing, and constructing morphological, part-of-speech and semantic class features; representative keywords are extracted by using a YAKE algorithm combined with part-of-speech filtering to form a candidate feature set; morphological information is input into a pre-trained language model to obtain context semantic representation, part-of-speech and semantic class information are input into a static word vector model to obtain language knowledge representation, and the two types of representation are input into a long sequence coding model after being fused to capture long-distance semantic dependence; then, the multi-source features are weighted and fused through a self-attention mechanism to obtain target text representation; in the model training process, difficult samples are screened based on prediction confidence and sample loss value, the difficult samples are subjected to data enhancement, the enhanced samples and original training samples are combined to form a new training sample set, and the model is trained in combination with cross-entropy loss and focal loss; finally, the word sense class of an ambiguous word in the current context is determined according to the classification result. The application can reduce the interference of redundant features, improve the accuracy and robustness of word sense disambiguation in complex contexts, and is especially suitable for semantic understanding tasks in professional fields such as biomedical texts.

The application discloses a similarity discrimination method and system for new word sense primitive recommendation, and comprises the following steps: in the HowNet word set, similar words to a new word are selected by a similarity discrimination model to form a similar word set; a local "word-semantic item-semantic primitive" relationship network is constructed according to all the words in the similar word set, the corresponding concept semantic items and semantic primitives of the words; semantic primitive nodes are selected based on a network node importance sorting method, a recommendation index of the semantic primitive nodes is generated according to the standardization degree centrality and the intermediate centrality of the semantic primitive nodes, the importance of the semantic primitive nodes is evaluated, the correlation between an unregistered word and a semantic primitive is established by taking the similar word set as a bridge, the sorting and selection of the candidate semantic primitives of the unregistered word are completed through the recommendation index, and the HowNet is expanded through the new word; the application effectively solves the similarity discrimination problem between the unregistered word and the word table word, and effectively solves the selection problem of the candidate semantic primitives.

The present invention discloses a long-tail word sense disambiguation method integrating decoupled representation, including: learning the word embedding of the target word from the text to be disambiguated, that is, the target word embedding, where the text-to-vector mapping model is implemented by a target word encoder; learning the text embedding of the word sense definition from the word sense definition text in the dictionary, that is, the word sense definition embedding, where the text-to-vector mapping model is implemented by a definition encoder; duplicating the obtained target word embedding and word sense definition embedding, one directly used to calculate the similarity score of the word sense under the traditional representation method; the other is reshaped by the decoupled representation method to obtain the similarity score of the word sense under the decoupled representation, and finally the scores under the two representation methods are weighted and summed as the output value; the decoupled representation method is a representation method inspired by the entangled state in quantum theory, based on the VAE model framework, and can effectively reduce the sampling noise of the original VAE model.

The present invention discloses a long-tail word sense disambiguation method integrating a definition and a scenario matching mechanism, including: learning a text embedding of a word sense definition (i.e., a definition embedding) from a word sense annotation text in a dictionary, wherein a text-to-vector mapping model is implemented by a definition encoder; learning a text embedding of a word sense scenario (i.e., a scenario embedding) from a word sense example sentence text in the dictionary, wherein a text-to-vector mapping model is implemented by a scenario encoder; in the process of obtaining the scenario embedding, hiding the target word contained in the input text to obtain the scenario information of the word sense; learning a word embedding of the target word (i.e., a target word embedding) from a text to be disambiguated containing the target word; the pre-trained language model for obtaining the target word embedding and the pre-trained language model for obtaining the scenario embedding share one; and superimposing and summing the probabilities of the calculation results under the definition matching mechanism and the scenario matching mechanism as the final output value to determine the final word sense of the target word.

Aiming at the defects of an existing test method in the field of word sense disambiguation test of machine translation in the aspects of triggering effective word sense conversion, maintaining original sense and maintaining semantic consistency, the invention provides a machine translation polysemous word translation evaluation method based on semantic item trigger word replacement. The method comprises the following steps: firstly, constructing a polysemy semantic item library, and screening source language sentences containing polysemy from a parallel corpus according to the semantic item library; performing natural language processing on the sentence, and identifying a trigger word of a polysemy special definition item in the sentence; replacing the trigger word with a replacement word to generate a variant sentence; performing machine translation and alignment on the original sentences and the variant sentences to obtain expressions of polysemy words in translations; and calculating the similarity of the polysemy translation expression, and judging whether a polysemy disambiguation error exists or not. Through a trigger word replacement strategy, a semantic controlled contrast sentence pair can be constructed, potential errors of a system in polysemous word translation are effectively revealed, and the pertinence and effectiveness of evaluation are improved.

The invention provides a traditional Chinese medicine cross-language aided translation system based on artificial intelligence, and belongs to the technical field of traditional Chinese medicine translation. The traditional Chinese medicine cross-language aided translation system based on artificial intelligence comprises a data acquisition module, a semantic recognition module, a semantic correction module, a translation strategy output module and a translation evaluation module. The system can track semantic distance changes between term translation and historical standards in real time, a translation migration file is established for recognized deviations, the corresponding relation between word meaning drift paths and contexts is recorded, the system can judge whether the deviations belong to common error types or not, rapid error correction is conducted on the basis of a term evolution database, and therefore the accuracy of term translation is improved. The mechanism does not depend on a fixed dictionary rule, but performs combined judgment according to factors such as a semantic change trend, a context structure adaptation degree and the frequency of historical deviation behaviors, so that the intelligence and the adaptability of deviation judgment are greatly improved, and the self-correction capability and the fault tolerance of a translation system are remarkably improved.

The application discloses a large model knowledge graph representation method based on a Transformer, and comprises the following steps: (1) randomly sampling a subgraph containing a central triple from a knowledge graph to construct a mask subgraph sequence; (2) extracting embedding representation of nodes in the mask subgraph sequence by using the Transformer, wherein a multi-dimensional power of an adjacency matrix is used as structure information during extraction, and an encoding vector of the structure information is added to an attention mechanism of the Transformer to obtain embedding representation of a mask node; (3) performing word sense prediction on the embedding representation of the mask node by using a classifier to obtain a word sense prediction result; (4) constructing a loss function and optimizing model parameters; and (5) performing completion of the knowledge graph by using the model with optimized parameters. The method can fully capture structure information and context semantic information in the knowledge graph.

The application discloses an event extraction system and method based on a pre-training model and word sense enhancement, a domain word vector acquisition module obtains a domain word vector; a data set construction module obtains a specific domain text sequence data set; an event extraction model construction module constructs an event extraction model; a training module takes the domain word vector as an initial setting for training in a Soft-lexicon sub-model in the event extraction model, and trains the event extraction model by using the labeled specific domain text sequence data set; and an event extraction module predicts a label sequence result of the specific domain text sequence data set to be detected by using the trained event extraction model. By using the pre-training LERT sub-model and the Soft-lexicon sub-model, the application incorporates the domain vocabulary information of the domain word vector into the character representation, introduces a CRF layer in the model, improves the accuracy of label annotation in the event extraction task, and solves the problems of data scarcity, strong professionalism and complex context in specific domain event extraction.

The present disclosure provides an intention recognition method, an electronic device, a readable storage medium and a computer program product. The intention recognition method of the present disclosure comprises: extracting word sense information and word boundary information of characters in the recognized text; converting the word sense information and the word boundary information into a plurality of probability distribution vectors based on a self-attention algorithm, the probability distribution vectors representing the probability of the predicted characters appearing in the predicted sequence; screening at least one candidate vector sequence from the plurality of probability distribution vectors, the candidate vector sequence being a probability distribution vector whose relevance to any prototype vector sequence in a prototype vector sequence library is higher than a relevance threshold; and generating an intention summary text based on the candidate vector sequence.