Open AccessNavy text data set equalization method based on deep learning
Author(s) -
Yongsheng Qi,
Ding Haiqiang,
Jinchao Zhao
Publication year - 2019
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1345/2/022013
Subject(s) - computer science , set (abstract data type) , data set , artificial intelligence , navy , text processing , data mining , natural language processing , universality (dynamical systems) , training set , information retrieval , programming language , archaeology , history , physics , quantum mechanics
Traditional unbalanced data sets, such as artificial set processing method, exists in the complicated and poor universality. It is difficult to apply naval uneven text data set processing. Aiming at this problem, this paper proposes a Navy uneven text data processing model based on biRNN. It learns text sequence features through biRNN model. Then it generates text similar to the original to balance text data. Text classification experiments are carried out on original data set and balanced data set respectively. The experimental results show that balancing the original data set based on biRNN method can effectively improve the performance of text classification.