AGCNN-Sentence-Classification

Demo code for the paper An Attention-Gated Convolutional Neural Network for Sentence Classification
(accepted for publication in the IDA journal).
Runs the AGCNN-rand model on the CR dataset (Customer Review dataset, by M. Hu and B. Liu).

Requirements

• Python (3.6)
• Tensorflow (1.12.0_gpu)
• Gensim (3.7.3)
• Pre-trained word2vec google word vectors https://code.google.com/archive/p/word2vec/

Note

Usage of this code may need to modify some statements, such as file path.
If the code is helpful for your work, please cite our paper as
"Yang Liu, Lixin Ji, Ruiyang Huang, Tuosiyu Ming, Chao Gao, and Jianpeng Zhang (2019), An Attention-Gated Convolutional Neural Network for Sentence Classification, Intelligent Data Analysis (in press)."
or
"Yang Liu, Lixin Ji, Ruiyang Huang, Tuosiyu Ming, Chao Gao, and Jianpeng Zhang (2018), An Attention-Gated Convolutional Neural Network for Sentence Classification, arXiv preprint arXiv:1808.07325."