NORMA eResearch @NCI Library

Topic Classification: Hybrid Feature selection model using BPSO-MLP

Kunchum Satheesh, Karthikranjan (2018) Topic Classification: Hybrid Feature selection model using BPSO-MLP. Masters thesis, Dublin, National College of Ireland.

[thumbnail of Master of Science]
Preview
PDF (Master of Science)
Download (3MB) | Preview

Abstract

In this 21st century era of information and technology where our day to day life is filled with online activities which generate data by leaps and bounds. Majority of data being generated is of natural language/ text data format which becomes difficult to store. if this data is managed and analyzed properly has immense potential to make businesses more convenient. So, one must provide a way through to navigate and process this text data of reports for categorising the topics of interest. Since it would be a challenging task to handle high dimensional data and filter a relevant information, feature selection algorithms were used to enhance the efficiency of the topic classification and improve in a choice of selecting significant information. This study investigates the efficiency of the hybrid feature selection technique based on binary particle swarm optimization and evaluated with multi-layered perceptron to determine the quality of features. Experimental results show that the combination of multi-layered perceptron with B-PSO resulted in a good accuracy rate of 83% with a reduced number of features.

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: Caoimhe Ní Mhaicín
Date Deposited: 06 Nov 2018 12:35
Last Modified: 06 Nov 2018 12:35
URI: https://norma.ncirl.ie/id/eprint/3447

Actions (login required)

View Item View Item