Text Classification of Enterprise Technical Requirements Based on RCNN_ATT Model

DOI

10.2991/assehr.k.210513.118How to use a DOI?

Keywords

Text classification, Attention mechanism, Long and short term memory, Max pooling, Technical requirements

Abstract

The text information of enterprise technical requirements is miscellaneous, which leads to the feature extraction is not prominent enough, and cannot be further accurately and effectively matched to the scientific research team of colleges and universities. In this paper, attention mechanism is added to the two way LSTM network to calculate the contribution score of the category to which the output vector belongs, and the word vector combined with attention matrix is connected to the maximum pooling layer, then RCNN_ATT model for enterprise technical requirement text is proposed, so that the technical requirements text can be automatically classified according to the industry. The experimental results show that, compared with other neural network models, this model performs better in technical requirement text classification, which can narrow the scope of supply and demand matching and improve the efficiency of matching calculation.

Copyright

© 2021, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Xingbing Liu
AU  - Bin Chai
AU  - Yingying Wang
AU  - Yachao Zhai
PY  - 2021
DA  - 2021/05/14
TI  - Text Classification of Enterprise Technical Requirements Based on RCNN_ATT Model
BT  - Proceedings of the 6th International Conference on Education Reform and Modern Management (ERMM 2021)
PB  - Atlantis Press
SP  - 513
EP  - 519
SN  - 2352-5398
UR  - https://doi.org/10.2991/assehr.k.210513.118
DO  - 10.2991/assehr.k.210513.118
ID  - Liu2021
ER  -