Deep learning for sentence/text classification
Deep Learning Architectures have been achieving state-of-the-art results in many application scenarios. Particularly, the performance of Deep Convolution Neural Networks (Deep ConvNets) in computer vision tasks is incontestable. The wave of ConvNets is sweeping through other applications other than...
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Theses and Dissertations |
| Language: | English |
| Published: |
2018
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/10356/76043 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Nanyang Technological University |
| Language: | English |
| Summary: | Deep Learning Architectures have been achieving state-of-the-art results in many application scenarios. Particularly, the performance of Deep Convolution Neural Networks (Deep ConvNets) in computer vision tasks is incontestable. The wave of ConvNets is sweeping through other applications other than vision tasks. There are some instances of ConvNets used for Natural Language Processing (NLP) tasks such as sentence/text classification. The objective of this project is applying Deep Learning models such as Recurrent Neural Networks, ConvNets for sentence/text classification tasks and suggest ways to improve their performance. In this design, I used CNN(Convolution neural network) network structure as my framework, using python3 programming language and PyTorch deep learning tool to complete the preparation of the software and experiments on the remote server in the laboratory to get the final result(using GPU acceleration). |
|---|