Detection of emotion by text analysis using machine learning

Area of Science:

• Artificial Intelligence
• Natural Language Processing
• Computational Linguistics

Background:

• Human emotions are complex and challenging for machines to understand.
• Accurate emotion detection is crucial for improving human-machine interaction, especially in conversational agents like chatbots.
• Existing methods struggle with the nuances of emotional expression in text.

Purpose of the Study:

• To develop and compare machine learning models for automatic emotion detection in text.
• To enable machines, such as chatbots, to accurately assess human emotional states.
• To enhance human-machine communication by adapting to detected emotions.

Main Methods:

• Experiments utilized a lexicon-based approach and classic machine learning methods (Naïve Bayes, Support Vector Machine).
• Deep learning models, specifically neural networks, were employed for text-based emotion detection.
• Model effectiveness was evaluated using a multi-classification task across six emotions.

Main Results:

• Neural network models demonstrated superior performance, achieving an F1-score of 0.95 for sadness.
• The developed emotion detection model was successfully integrated into a web application for text analysis.
• The model enhanced chatbot communication by providing real-time emotional state insights.

Conclusions:

• Machine learning approaches, especially neural networks, show significant potential for accurate emotion detection from text.
• The research validates the practical application of emotion detection models in improving human-machine interaction.
• Full automation of emotion detection remains an open research area requiring further advancements.