Neighbor analysis-combined named entity identification method based on two-stage classification
A named entity recognition, entity technology, applied in character and pattern recognition, instruments, computing and other directions, can solve the problems of blurred boundaries, multiple data types in datasets, and decreased entity recognition performance.
Active Publication Date: 2022-06-03
KUNMING UNIV OF SCI & TECH
View PDF17 Cites 2 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0005] The present invention provides a named entity recognition method combined with two-stage classification of neighbor analysis, which is used to classify fine-grained entities, and solves the problem of Entity recognition performance degradation caused by multiple categories and blurred boundaries
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment 1
[0048] Example 1: figure 1 Show, a two -stage category named entity recognition method combined with near -neighborhood analysis. The specific steps of the method are as follows:
[0049] STEP1, pre -process the English text and labels obtained from the Chinese naming data set from CLUENER2020, take out each sentence in the text, the number corresponding to the entity in each sentence, the position of each entity in the sentence, and the position of each entity in the sentence, and the position of each entity in the sentence, and the position of each entity in the sentence, and the position of each entity in the sentence, and the position of each entity in the sentence, and the position of each entity in the sentence, and the position of each entity in the sentence, and the position of each entity in the sentence, and the position of each entity in the sentence, and the position of each entity in the sentence, and the position of each entity in the sentence, and the position of ea...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention provides a two-stage classification named entity recognition method in combination with neighbor analysis, belongs to the field of natural language processing, and provides a method for combining neighbor analysis and entity two-stage classification for the first time in order to solve a named entity recognition task. The problems of multiple categories, fuzzy entity boundary and the like existing in a fine-grained entity classification data set can be well solved. According to the method, the training set is used for training the entity range detection model to obtain the optimal entity range detection model, then the optimal entity range detection model is used for detecting the to-be-recognized data set, the to-be-recognized data set with the entity ranges is obtained, the feature vectors corresponding to the head and tail positions of each entity range are spliced to serve as input of the entity classifier, and therefore the recognition accuracy of the to-be-recognized data set is improved. The whole model is adjusted by using the real entity category, so that the model has better generalization ability, and finally, the recognition effect of the model on the data set to be recognized is improved.
Description
Technical field [0001] The invention involves a two -stage classification method of combining neighborhood analysis, which is the field of natural language processing technology. Background technique [0002] Natural language processing is an important direction in the field of computer science and artificial intelligence. Study various theories and methods for effective communication between people and computers with natural language. Natural language treatment is mainly used in machine translation, public opinion monitoring, automatic abstract, viewpoint, text classification, question answering, text semantic comparison, voice recognition, etc. Natural language treatment usually includes word division, word marking, naming entity recognition and grammar analysis. Naming entity recognition is the basic task of information extraction in natural language processing. It aims to locate and classify naming entities from non -structureized natural languages. Named entity recognition i...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Applications(China)
IPC IPC(8): G06F40/295G06K9/62
CPCG06F40/295G06F18/24147G06F18/2415
Inventor 线岩团江汀莹余正涛
Owner KUNMING UNIV OF SCI & TECH