NORMA eResearch @NCI Library

Text Classification using Graph Based Learning

Ghosh, Soumyadip Dipak (2020) Text Classification using Graph Based Learning. Masters thesis, Dublin, National College of Ireland.

[img]
Preview
PDF (Master of Science)
Download (1MB) | Preview
[img]
Preview
PDF (Configuration manual)
Download (311kB) | Preview

Abstract

Current text classification models based on deep learning increasingly rely on local word occurrence information and sequential semantics. These techniques the information residing long distance semantics and global word occurrence information. Recently, numerous researchers have explored graph neural networks (GNN) to apply on plethora of tasks as graph structures are particularly adept in capturing complex and abstract relations between entities which may prove to be highly useful in natural language processing related tasks. Convolution neural networks (CNN) have been very effective in deep learning for highly structured data like texts and images. Thus studies have recently begun to harness the power of convolution in a manner which would be effective for application on graph-structured data with GNNs. However, very less studies have explored the use of graph convolutional networks (GCN) for the purpose of text classification. This study aims to construct a graph from a corpus of text comprising of documents and words as nodes and use it for text classification using GCN. This will enable the neural network to learn from complex information residing in relationships between document-word and word-word co occurrences. The approach shows state-of-the-art performance in multi-label classification on two out of the four popular benchmarking corpora used in this work to test our approach.

Item Type: Thesis (Masters)
Subjects: Q Science > QA Mathematics > Electronic computers. Computer science
T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science

Q Science > QA Mathematics > Computer software
T Technology > T Technology (General) > Information Technology > Computer software
Divisions: School of Computing > Master of Science in Data Analytics
Depositing User: Dan English
Date Deposited: 22 Jan 2021 12:41
Last Modified: 22 Jan 2021 12:41
URI: http://norma.ncirl.ie/id/eprint/4443

Actions (login required)

View Item View Item