Named entity recognition method for field of power planning review
A named entity recognition and planning technology, which is applied in neural learning methods, electrical digital data processing, instruments, etc., can solve the problems of unsatisfactory automatic recognition and the inability of staff to name entity recognition, so as to improve recognition ability, efficiency, and good The effect of precision and recall
Pending Publication Date: 2022-03-08
STATE GRID SICHUAN ECONOMIC RES INST
View PDF0 Cites 1 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0005] The technical problem to be solved by the present invention is that the automatic identification of named entities in the current power grid planning field is not ideal, and in most cases it is still necessary to rely on manual means for entity extraction, especially when the text corpus is completely unstructured text, the existing extraction method No longer applicable, resulting in the inability of current staff to perform effective named entity recognition work in unstructured power grid planning text corpora
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
[0058] Such as figure 1 As shown, the present invention provides a named entity recognition method oriented to the field of electric power planning review, which mainly includes the following steps:
[0059] Step S1: According to the characteristics of the sample corpus data set, set up multiple entity labels that describe the different properties of entities, and obtain the sample corpus after word segmentation;
[0060] Step S2: Load the sample corpus after word segmentation to the Glove model to train word vectors, and splicing by position to obtain a text sequence vector matrix
[0061] Step S3: Use the multi-scale convolutional network to process the text sequence vector matrix After convolution, recombine and pool to extract the lexical information of the word granularity in the sequence;
[0062] Step S4: Convert the text sequence vector matrix Concatenate the end states of forward LSTM and backward LSTM in the input BiLSTM network, and extract the affix informat...
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 discloses a named entity recognition method for the field of power planning review, which comprises the following steps: S1, according to the characteristics of a sample corpus data set, setting a plurality of entity tags for describing different properties of entities, and obtaining a sample corpus after word segmentation; s2, loading the sample corpora subjected to word segmentation to a Glove model to train word vectors, and splicing according to positions to obtain a text sequence vector matrix; s3, the text sequence vector matrix is recombined and pooled after being subjected to convolution through a multi-scale convolution network, and word information of word granularity in the sequence is extracted; s4, splicing end states of a forward LSTM and a backward LSTM in the BiLSTM network taking the text sequence vector matrix as input, and extracting affix information of sentence granularity in the sequence; and S5, fusing the vocabulary information of the word granularity in the sequence and the affix information of the sentence granularity in the sequence by using a Cross-Transform module, and finally completing named entity recognition through a CRF layer. According to the method, the named entity recognition efficiency in the electric power planning review field is improved to a certain extent.
Description
technical field [0001] The invention relates to the fields of power planning and computer technology, in particular to a named entity recognition method oriented to the field of power planning review. Background technique [0002] With the rapid development of computer technology in modern society, coupled with the widespread use of artificial intelligence technology and software technology, the mode of artificially generating power grid planning review results can no longer meet the needs of power enterprise development and planning departments to improve the level of grid planning work, improve grid planning work efficiency and ensure Requirements for the quality of grid planning work. The traditional power grid planning method adopts the method of manually making excel tables for data management. This method not only has low work efficiency and high work intensity, but also has a low safety factor and is not easy to save. It is easy to cause power grid data to be leaked a...
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): G06F40/295G06N3/04G06N3/08
CPCG06F40/295G06N3/08G06N3/044
Inventor 罗劲瑭姚实颖冯渝荏徐杰杨宇玄陈一鸣曾鉴祝和春余葭苇倪江张晨琳
Owner STATE GRID SICHUAN ECONOMIC RES INST