Improving Emotion Detection in Text: A Comparative Analysis of Machine Learning Algorithms and Genetic Algorithm-Optimized Logistic Regression
K. Hemakirthiga, J. Arunadevi · 2023
Emotion detection, a critical task in natural language processing, plays a vital role in understanding and analyzing emotions expressed in text data. This study focuses on Emotion detection of the Tweet_emotions dataset, a challenging multi-class classification problem with 13 emotion classes. We compare the performance of various classifiers, including logistic regression, random forest, SVM, gradient boosting, Naive Bayes, and KNN. We suggest using a genetic algorithm (GA) optimization technique for tweaking the logistic regression classifier’s hyperparameters in order to improve its performance. The experimental results demonstrate that the GA-optimized logistic regression classifier achieved the highest accuracy of 90.07%, outperforming the other classifiers. Our findings contribute to the field of Emotion detection by showcasing the effectiveness of the GA optimization technique in enhancing the performance of logistic regression for multi-class sentiment classification tasks. This research has significant implications for understanding emotions expressed in textual data and can be applied in various domains such as social media analysis, customer feedback analysis, and opinion mining.