Sentimental analysis from imbalanced code-mixed data using machine learning approaches.

Area of Science:

• Natural Language Processing
• Machine Learning
• Computational Linguistics

Background:

• Sentiment analysis is crucial for knowledge discovery across fields.
• Class imbalance is a significant challenge in sentiment analysis, particularly with code-mixed data.
• Existing research has largely overlooked sentiment analysis in imbalanced, code-mixed datasets.

Purpose of the Study:

• To address the challenge of class imbalance in sentiment analysis for code-mixed text.
• To propose and evaluate a novel approach for sentiment analysis on imbalanced code-mixed data.
• To compare the effectiveness of various machine learning classifiers for this specific task.

Main Methods:

• A combination of sampling techniques and Levenshtein distance metrics was employed.
• The study evaluated multiple machine learning algorithms: Random Forest, Logistic Regression, XGBoost, Support Vector Machine, and Naïve Bayes.
• Performance was assessed using the F1-Score metric.

Main Results:

• The proposed method effectively handles class imbalance in code-mixed sentiment analysis.
• Comparative analysis revealed the performance variations among different machine learning classifiers.
• The F1-Score was utilized to quantify and compare the effectiveness of the implemented approaches.

Conclusions:

• The developed approach offers a viable solution for sentiment analysis in challenging code-mixed, imbalanced datasets.
• The findings provide insights into the suitability of different machine learning models for this task.
• Further research can build upon these methods to enhance sentiment analysis accuracy in multilingual contexts.