Hyperparameter Optimization of Semi-Supervised Sentiment Annotation Model on Marketplace Dataset

DOI

10.2991/978-2-38476-247-7_18How to use a DOI?

Keywords

hyperparameter tuning; machine learning; sentiment analysis

Abstract

Hyperparameter optimization in semi-supervised learning (SSL) models for sentiment analysis is the process of adjusting the machine learning model’s parameters to enhance the performance of the SSL system. Hyperparameter tuning aims to maximize the accuracy, precision, recall, and F1-Score of each machine learning component in the SSL model. The SSL model subjected to hyperparameter tuning is an outcome of previous research specifically designed for the automatic annotation process in sentiment analysis. This paper aims to enhance the performance of machine learning within the SSL model, which includes the Support Vector Machine (SVM) algorithm and Random Forest classifier. The parameter in the Random Forest classifier is the number of trees (estimators). For the SVM method, the parameter in question is the value of ‘C.’ The hyperparameter testing employed the Random Search method. The dataset used consists of customer comments from a marketplace. The final result is hyperparameter tuning has succeeded in improving the sentiment annotation processing capabilities of SSL on the marketplace dataset. Improving the performance of sentiment annotation can be done by adjusting the C parameter in the SVM method and the number of estimator parameters in the Random Forest (RF) Classifier.

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  - Nur Heri Cahyana
AU  - Yuli Fauziah
AU  - Wisnalmawati
AU  - Agus Sasmito Aribowo
PY  - 2024
DA  - 2024/09/04
TI  - Hyperparameter Optimization of Semi-Supervised Sentiment Annotation Model on Marketplace Dataset
BT  - Proceedings of the 2nd International Conference on Advance Research in Social and Economic Science (ICARSE 2023)
PB  - Atlantis Press
SP  - 161
EP  - 170
SN  - 2352-5398
UR  - https://doi.org/10.2991/978-2-38476-247-7_18
DO  - 10.2991/978-2-38476-247-7_18
ID  - Cahyana2024
ER  -