Task Submission: AI Registration Assistant

Student: Abhinav Singh

Student Code: DAS002971

Domain: AI & Data Science

Internship ID: 760

Task ID: AI-SS-001

Project Overview

Task Title: AI Registration Assistant

The AI Registration Assistant is a conversational chatbot developed to help students understand and complete the internship registration process.

The chatbot can answer internship-related questions, collect student details, validate user input, guide users through registration steps and generate a registration confirmation.

Domain Summary

This project belongs to the AI and Data Science domain and focuses on Natural Language Processing, intent recognition, entity extraction and conversational AI.

The system understands user messages, identifies their intent and responds according to the current conversation state.

Technologies Used

• Python
• Flask
• NLTK
• Scikit-learn
• TF-IDF Vectorization
• Logistic Regression
• HTML
• CSS
• JavaScript
• JSON

Features Implemented

• Greeting and introduction
• Internship FAQ handling
• Intent recognition using Machine Learning
• Text preprocessing using NLTK
• Name and email entity extraction
• Email and input validation
• Multi-step registration process
• Dialog-state management
• Registration confirmation
• Unique registration ID generation
• JSON-based registration storage
• Basic Hindi and Hinglish support
• Sentiment analysis
• Chat interaction logging
• Responsive web interface
• Admin analytics dashboard

How I Completed This Task

I first planned the chatbot conversation flow and created different user intents, such as greeting, registration, eligibility, fees, help and task-submission information.

I used NLTK for basic text preprocessing. TF-IDF Vectorization was used to convert user messages into numerical features, and Logistic Regression was used to classify the user’s intent.

I implemented entity extraction and validation for details such as name and email address. Dialog-state management was used to track the current registration step.

After collecting all required information, the chatbot displays the entered details and asks the user for confirmation. Once confirmed, the registration data is stored in a JSON file and a unique registration ID is generated.

I also created a Flask-based web interface and an admin dashboard to display registration totals, chatbot interactions, intent distribution, sentiment information and recent registrations.

Challenges Faced

One of the main challenges was maintaining the correct registration flow while also allowing the chatbot to answer general questions.

I also faced compatibility issues while installing Scikit-learn with Python 3.14. I resolved this by creating the virtual environment using Python 3.13, which supports the required package versions.

Another challenge was validating invalid email addresses and ensuring that incomplete information was not saved.

What I Learned

Through this project, I learned:

• How NLP preprocessing works
• How TF-IDF converts text into numerical features
• How Logistic Regression performs intent classification
• How regular expressions help with entity extraction
• How input validation improves chatbot reliability
• How dialog-state management controls conversations
• How Flask connects Python with a web interface
• How JSON can be used for lightweight data storage
• How an admin dashboard can display chatbot analytics
• How to test and debug a complete AI application

Project Links

GitHub Repository:

https://github.com/ab9010/AI-Registration-Assistant-DAS002971

YouTube Demonstration:

https://youtu.be/Rnes_B_VrJU

Outcome

The AI Registration Assistant was completed successfully with all the main project requirements implemented.

The final application can understand common student queries, validate registration information, complete the registration workflow, store data in JSON format and display project analytics through an admin dashboard.

This project gave me practical experience in Natural Language Processing, Machine Learning, Flask development, chatbot design and frontend-backend integration.