Chatbot for College Enquiry System

Chatbot for College Enquiry System

Prospective students have numerous questions about colleges. This project builds an AI-powered chatbot for college enquiry. NLP tokenizes input, performs intent classification using TF-IDF cosine similarity against a knowledge base. The knowledge base covers courses, admissions, fees, facilities, placements. Deployed via Flask web application with a clean chat interface.

Components

Python 3.8+
NLTK/spaCy
Scikit-learn
Flask
SQLite

Components	Hexkart	Flipkart
Python 3.8+
NLTK/spaCy
Scikit-learn
Flask
SQLite

Note: The components listed are software tools and frameworks required for development.

NLP and Intent Classification

The chatbot processes user input through NLTK/spaCy for tokenization, part-of-speech tagging, and named entity recognition. The preprocessed query is converted to TF-IDF vectors. Intent classification uses cosine similarity to match the user query against the knowledge base entries, finding the most relevant predefined response. This allows the chatbot to correctly route questions about courses, admissions, fees, placements, and facilities to the right answers.

Knowledge Base

SQLite stores the structured knowledge base with categorized Q&A pairs. Categories include Courses (BTech, MTech, etc.), Admissions (process, eligibility, dates), Fee Structure (tuition, hostel, miscellaneous), Facilities (library, labs, sports), Placements (companies, statistics, training), and General Info (location, contact, events). The admin panel allows staff to add, edit, and remove Q&A entries, keeping the chatbot up to date with new courses, changed fees, and updated placement statistics.

Flask Chat Interface

The chatbot is deployed as a Flask web application with a modern, responsive chat interface.
Key chat features:

• AJAX-powered real-time messaging without page reloads.
• Typing indicator and message timestamps for natural feel.
• Fallback response when no match found, with query logging.
• Admin dashboard for reviewing unanswered questions and updating KB.

System Architecture

Key Functionalities

Chatbot Engine

• NLP text preprocessing: tokenization, stop word removal, normalization
• TF-IDF vectorization for converting queries to numerical features
• Cosine similarity calculation for matching queries to knowledge base
• Confidence threshold to trigger fallback for unknown queries

Knowledge Base Management

• Categorized Q&A pairs stored in SQLite database
• Admin panel for adding, editing, and removing entries
• Support for multi-language queries with translation integration
• Knowledge base expansion from logged unanswered queries

User Features

• Clean chat interface with message bubbles and timestamps
• Quick reply buttons for common query categories
• Option to escalate to human support for complex queries
• Chat history and session persistence