Enhancing Semi-Supervised Sentiment Analysis Through Hyperparameter Tuning Within Iterations: A Comparative Study Using Grid Search and Random Search

DOI

10.2991/978-94-6463-366-5_23How to use a DOI?

Keywords

Hyperparameter Tuning; Semi-supervised Learning; Sentiment Analysis

Abstract

Improper placement of hyperparameter tuning in the semi-supervised sentiment analysis process can potentially decrease processing speed and fail to enhance performance. This research explores the impact of hyperparameter tuning within iterations of semi-supervised sentiment analysis. Two architectural approaches are tested: one with hyperparameter tuning at the beginning and another with tuning at each iteration. Grid search and random search are employed for hyperparameter tuning. The study demonstrates that hyperparameter tuning within iterations enhances the performance of semi-supervised sentiment analysis models. The experiments conducted on four diverse datasets demonstrated that hyperparameter tuning within iterations generally leads to improved performance. Model B, which applies hyperparameter tuning within iterations, showed better accuracy, precision, recall, and F1-score than Model A, which conducts tuning at the beginning. Additionally, grid search outperformed random search, although the differences in performance were not highly significant, approximately from 0.1% to 2% in all experiments. These results suggest that incorporating hyperparameter tuning within the iterations of semi-supervised sentiment analysis can enhance model performance, and grid search can be a more effective method for this task, especially when time efficiency is a priority. The choice between grid and random search depends on the trade-off between time and performance. Future research can extend these findings to different machine learning techniques and datasets.

Copyright

© 2024 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Agus Sasmito Aribowo
AU  - Nur Heri Cahyana
AU  - Yuli Fauziah
PY  - 2024
DA  - 2024/02/02
TI  - Enhancing Semi-Supervised Sentiment Analysis Through Hyperparameter Tuning Within Iterations: A Comparative Study Using Grid Search and Random Search
BT  - Proceedings of the 2023 1st International Conference on Advanced Informatics and Intelligent Information Systems (ICAI3S 2023)
PB  - Atlantis Press
SP  - 248
EP  - 260
SN  - 1951-6851
UR  - https://doi.org/10.2991/978-94-6463-366-5_23
DO  - 10.2991/978-94-6463-366-5_23
ID  - Aribowo2024
ER  -