Context-Aware Hybrid Conversational Agent for College Enquiries Using Rule-Based and Machine Learning Approaches

In recent years, the use of conversational agents has increased in the education sector to simplify access to institutional information. However, most college enquiry chatbots remain limited by their reliance on static, rule-based logic and lack contextual understanding of user queries. This paper presents the design and implementation of a context-aware hybrid conversational agent specifically developed to handle college enquiry scenarios. The proposed system integrates both rule-based pattern matching and machine learning-based intent classification to create a robust, adaptive chatbot capable of handling both structured and unstructured queries. To achieve context-awareness, the chatbot tracks and retains recent conversation history within user sessions, allowing it to interpret follow-up or incomplete questions meaningfully. For instance, the chatbot can resolve queries like “What about the fee?” by linking them to the previously mentioned course in the conversation. A lightweight machine learning model trained on custom college enquiry datasets is used to classify user intent when the input does not match predefined rules. The hybrid approach ensures high accuracy for frequent questions while maintaining flexibility for varied or ambiguous queries. The system is deployed as a web-based application using Flask and can be easily adapted for other educational institutions. Through practical evaluation and test cases, the chatbot demonstrates improved response accuracy and a more natural user experience compared to traditional rule- based bots. This work contributes toward intelligent automation in educational support systems and highlights the importance of combining multiple AI techniques for effective conversational interfaces.