Development of emotion detection system using bidirectional long short-term memory networks.
Date
2024-11
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Federal University of Technology, Owerri
Abstract
Emotion detection in the context of chatbots holds immense promise for creating more empathetic and responsive conversational agents. This study presents a novel approach to enhancing chatbot capabilities by integrating emotion detection using Bidirectional Long ShortTerm Memory (BiLSTM) neural networks. The primary objective of this research is to equip chatbots with the ability to discern and adapt to the emotional states of users during interactions. Leveraging the advantages of BiLSTM,we develop a model that can capture the temporal dependencies and contextual nuances in user messages, enabling it to accurately identify emotions such as happiness, sadness, anger, and more. The chatbot's architecture is augmented with the emotion detection module, allowing it to continuously analyze user input and provide emotionally tailored responses. Through a comprehensive dataset of conversational exchanges enriched with emotional labels, our model is trained to understand the intricacies of emotional expressions within text. The results of our experiments demonstrate the efficacy of the BiLSTMbased emotion detection approach within the chatbot framework. Users experience more personalized and empathetic interactions as the chatbot adapts its responses to match the detected emotional states. Comparative evaluations against traditional rule-based and non-emotion-aware chatbots underscore the significant improvements in user engagement and satisfaction. In conclusion, this research represents a significant advancement in the field of conversational Artificial Intelligence (AI). The integration of BiLSTM-based emotion detection empowers chatbots to better understand and respond to users' emotions, enhancing user experiences across a range of applications, from customer support to mental health companions. This work paves the way for more emotionally intelligent and empathetic AI-driven conversations, ultimately improving the quality and effectiveness of human-computer interactions.
Description
Master’s thesis on "emotion detection system". It contains diagrams, tables and graphs.
Keywords
Emotion, bidirectional, LSTM, RNN, dataset, chatbots, Department of computer science
Citation
Amadi, C. O .(2023). Development of emotion detection system using bidirectional long short-term memory networks. {Unpublished Master's Thesis}, Federal University of Technology,Owerri