Abstract

GemBot is an advanced AI chatbot that combines intelligent conversational capabilities with real-time sentiment analysis. It leverages the Gemini Pro API for natural language understanding and TextBlob for sentiment detection, providing both contextual responses and emotional insights. All user interactions and sentiment data are stored in MongoDB Atlas for further analysis and reporting. This system not only facilitates engaging conversations but also helps monitor user sentiment, making it ideal for customer support, feedback systems, and sentiment-driven personalization.

Introduction

The increasing demand for intelligent conversational systems in areas such as customer service, education, and mental health support calls for chatbots that do more than just respond to queries. GemBot addresses this by integrating sentiment analysis into chatbot conversations, allowing the system to track emotional tone in real-time. Combining Gemini Pro's advanced language model with TextBlob's sentiment analysis capabilities, GemBot offers both intelligent dialogue and emotional insights, which are logged into MongoDB Atlas for further analysis and trend monitoring.

This dual capability enables businesses and researchers to understand both what users say and how they feel, enhancing both user experience and data-driven decision-making.

Methodology

System Architecture:

The system consists of the following core components:

Frontend Interface:

A web interface where users can interact with GemBot.
Displays chatbot responses and sentiment analysis results in real-time.
Backend Processing

Accepts user queries through Flask-based APIs.
Sends the query to Gemini Pro API for contextual response generation.

Runs sentiment analysis using TextBlob, which calculates:
Polarity: Determines if the sentiment is positive, neutral, or negative.
Subjectivity: Measures how subjective (opinion-based) the input is.

Database Storage:

All interactions, including user inputs, chatbot responses, and sentiment analysis results, are stored in MongoDB Atlas.
Each record is timestamped for future trend analysis.

Technology Stack:
Gemini Pro API: For AI-powered conversational capabilities.
TextBlob: For sentiment analysis.
Flask: For backend API development.
MongoDB Atlas: For cloud-based data storage.
Python: Core programming language.
HTML/CSS/JavaScript: For frontend.

Workflow:
User submits input via the interface.
Backend forwards the query to Gemini Pro.
Response from Gemini Pro is displayed to the user.
Input is analyzed by TextBlob for polarity and subjectivity.
Sentiment scores are shown to the user alongside the chatbot response.
The entire interaction, including sentiment data, is saved in MongoDB Atlas for future analysis.

Experiments

Results

Intelligent Responses: Gemini Pro delivered context-aware and fluent responses across diverse user queries.
Real-time Sentiment Detection: TextBlob effectively categorized inputs into positive, neutral, and negative sentiments.
Data Logging & Analytics: MongoDB Atlas successfully stored user inputs, chatbot responses, and sentiment scores, creating a comprehensive interaction log for further analysis.
User Experience: Real-time sentiment feedback enhanced user engagement by making interactions more transparent.

Conclusion

GemBot successfully combines conversational intelligence with emotional awareness, offering a unique chatbot solution that not only responds to user queries but also tracks how users feel. Its ability to log and analyze sentiment trends provides valuable insights for businesses to understand user emotions at scale. With future enhancements such as multilingual support, sentiment dashboards, and voice-based interactions, GemBot has the potential to become a comprehensive AI-powered emotional intelligence platform for businesses and researchers alike.