Shale gas entity recognition method based on improved neural network

Abstract

The invention provides a shale gas entity recognition method based on an improved neural network, and belongs to the field of shale gas and natural language processing. The method comprises the following steps: firstly, preprocessing shale gas manual annotation data, mapping characters into a dense vector sequence with context semantics, and transmitting the dense vector sequence to a convolutional neural network to filter the influence of local contexts in sentences on entity recognition; then, the hidden state of the context at the shale gas labeling position is captured through a bidirectional long-short-term memory network, labels in sentences are concerned through an attention mechanism, character labels are predicted through similar information, and the problem of labeling inconsistency is solved; and finally, uploading to a conditional random field, and further constraining the result to achieve the effect of entity classification. According to the method, the names of related entities in the shale gas field can be quickly and efficiently identified, and the first shale gas high-precision entity identification method is provided for a shale gas intelligent analysis system.

Description

technical field [0001] The invention relates to the field of shale gas and natural language processing, in particular to a shale gas entity identification method based on an improved neural network. Background technique [0002] With the accelerated pace of unconventional oil and gas exploration and development, in order to further deepen the informatization processing of shale gas data and build an intelligent analysis system, it is necessary to analyze and process the underlying data and extract corresponding physical objects. However, most of the traditional shale gas data analysis is to study structured data and use data warehouse tools for mining. Advanced Named Entity Recognition (NER) technology is not used. [0003] Named Entity Recognition (NER), as the key technology of semantic extraction, recognizes and classifies entity names in samples. In the years of NER research, most of them are for English texts, and a few Chinese NERs are used in some specific fields, b...

AI Technical Summary

Problems solved by technology

[0005] The starting point of the present invention is to overcome the deficiencies of the original technology, and provide a shale gas entity recognition method based on an improved neural network. This method solves the messy data structure in the shale gas field by introducing an attention mechanism, and there are a large number of abbreviations and abbreviations. The inconsistency of entity labels caused by incomplete text makes it more suitable for the shale gas field and becomes the first entity recognition method in the shale gas field

Images