Authors: Aditya Kamble, ,Jayash Chavan, Siddharth Gaikwad, Vaishnavi Dardige, Professor I.T. Mukharjee
Abstract: In recent years, conversational artificial intelligence (AI) has transitioned from simple command-based systems to emotionally responsive digital companions capable of contextual understanding and human-like interaction. However, most available chatbot architectures depend on large-scale machine learning models and cloud-based computation, making them complex, data-intensive, and inaccessible for lightweight or educational applications. To address these challenges, this research presents Buddy, a modular, Python-based conversational chatbot designed to deliver personalized entertainment and motivational support using emotion cues and real-time API integrations. The proposed framework emphasizes simplicity, scalability, and emotional adaptability without requiring extensive natural language processing infrastructure. Buddy functions as a context-aware conversational assistant capable of recognizing user moods and providing tailored responses in the form of movie, music, weather, and motivational recommendations. The system integrates multiple public APIs — including TMDb for film suggestions, Spotify for mood-aligned music, ZenQuotes for motivational content, and OpenWeatherMap for contextual awareness — enabling multi-domain interaction. A rule-based mood detection engine identifies user sentiment from keywords such as “sad,” “happy,” or “bored,” while modular functions handle data fetching and formatting. This architecture creates a seamless conversational loop where Buddy adapts to user feedback, learns preferences in real time, and maintains an empathetic conversational tone. Experimental evaluation demonstrates that Buddy achieves an average response latency of 1.25 seconds, with a keyword detection accuracy of 95% across varied moods and network conditions. The framework exhibits low computational overhead (<100 MB memory usage) and functions efficiently on standard hardware without GPU support. Furthermore, its modular structure allows integration with advanced AI components such as sentiment classifiers or edge-deployed models in future versions. Compared to existing AI chatbots, Buddy provides a balance between human-like engagement, data privacy, and system transparency, highlighting how modular API-driven design can democratize access to emotionally intelligent AI systems. This study establishes Buddy as a sustainable, privacy-preserving, and adaptable conversational framework that bridges the gap between rule-based logic and emotionally aware digital companionship — paving the way for next-generation lightweight AI assistants suitable for education, mental wellness, and personalized entertainment ecosystems.
