Swarm Intelligence for Sentiment Analysis from Comments
No Thumbnail Available
Date
2024
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Sentiment analysis is an NLP technique used to identify the emotional tone of a
piece of text. This method helps discern the sentiment conveyed in texts like reviews,
comments, social media posts, or any other data involving human language.
To understand and categorise the sentiment conveyed in textual data, machine
learning offers methods and tools necessary to create models capable of this. ML
techniques are numerous, for this work Support Vector Machine(svm) is used; but
this algorithm relies on a set of hyper-parameters which greatly influence its performance.
Therefore, this study takes hyper-parameter tuning as an optimization
problem and employs a swarm-based optimization technique to enhance the performance
of SVM in sentiment analysis by using the recently introduced Termite
Alate Optimization Algorithm. The performance of proposed approach is evaluated
using five metrics: accuracy, precision, recall, F1-measure and computation
time on five well known datasets in the field of sentiment analysis and compared
to another famous swarm-based algorithm: Particle Swarm Optimization. The
experimental results show significant improvement in the SVM performance with
optimized hyper-parameters by the proposed approach compared to SVM with
default hyper-parameters in all datasets. Remarkably, TAOA demonstrated faster
tuning compared to PSO, completing its process with notable efficiency.