This repository houses the implementation of a cutting-edge conversational bot designed to enhance the online shopping experience by providing insightful interactions based on product reviews. The system employs advanced natural language processing techniques to generate context-aware, engaging, and informative responses, making review navigation intuitive and user-friendly.
With the proliferation of e-commerce, consumers increasingly rely on product reviews to make purchasing decisions. However, the vast quantity of reviews can overwhelm users, making it difficult to discern valuable insights. Our project addresses this challenge by developing a conversational AI that synthesizes review data into meaningful interactions, thereby enhancing the decision-making process for online shoppers.
- Advanced NLP Capabilities: Utilizes RASA for intent classification and GPT-2 for generating conversational responses, ensuring interactions are both natural and relevant.
- Sentiment Analysis: Employs sentiment analysis to gauge the tone of reviews, allowing the bot to differentiate between positive and negative feedback effectively.
- Contextual Awareness: Maintains conversational context, enabling the bot to handle follow-up questions intelligently and maintain coherence throughout interactions.
The system integrates two main components:
- Intent Classification: Built with RASA, this component categorizes user inquiries into predefined intents, allowing the bot to understand and address various user queries efficiently.
- Conversational Agent: Powered by GPT-2, this component generates responses that are contextually relevant and engaging, based on the input from the intent classification system and the sentiment analysis results.
For a comprehensive understanding of the methodologies and insights developed in this project, please refer to our project report available here.