University Assistant Chatbot

A desktop chatbot application built with Python and Flet, designed to help university students quickly access course information, schedules, deadlines, and contact details. Features a clean chat interface with real-time responses.

Overview

This chatbot serves as a virtual assistant for University of Greenwich students, providing instant access to:

• Weekly class schedules with room numbers
• Assignment deadlines for all courses
• Professor contact information
• Course-specific queries

Built as a proof-of-concept for improving student access to academic information.

Features

• Interactive Chat Interface: Clean, modern UI with message bubbles
• Natural Language Processing: Understands casual queries like "when is english class?" or "contact for web design"
• Comprehensive Data Coverage: Includes schedules, deadlines, and contacts for 5 courses
• Real-time Responses: Instant answers without external API calls
• Desktop Application: Runs locally using Flet framework

Screenshots

Chat Interface

Schedule Query

Deadline Information

Contact Lookup

Tech Stack

Language: Python 3.10+
UI Framework: Flet (Python desktop UI framework)
Data Storage: In-memory dictionaries (for prototype)
Architecture: Event-driven chat interface

Installation & Setup

Prerequisites

• Python 3.10 or higher
• pip package manager

Install Dependencies

# Clone the repository
cd Chatbot-Flet

# Install required packages
pip install flet

Run the Application

python chatbot.py

chatbot.py is the main documented version of the app. The repository also includes main.py, an alternative earlier UI implementation with similar sample data.

A window will open with the chatbot interface ready to use.

Usage Examples

Get your weekly schedule:

User: schedule
Bot: [Displays full weekly timetable with times and rooms]

Find assignment deadlines:

User: deadlines for web design
Bot: Assignment due on 9th August: Develop a responsive website...
     Presentation due on 9th August: Present the developed website...

Get professor contact:

User: contact for english
Bot: Contact for English for Academic Purposes: Laura Laubacher - l.k.laubacher@greenwich.ac.uk

Check specific day schedule:

User: what's on thursday?
Bot: Thursday Schedule:
     - Computer Research and Technology from 9-11am at Room 601 (Lab L12)
     - Web Design and Development from 2-4pm at Room 601 (Lab L12)

Data Structure

The chatbot uses three main data dictionaries:

Schedules

schedules = {
    "Monday": "No classes",
    "Tuesday": [
        {"subject": "English for Academic Purposes", "time": "9-12pm", "room": "Room 105 (18)"},
        {"subject": "Project Study", "time": "12-2pm", "room": "Room 205 (Lab L23)"}
    ],
    # ... additional days
}

Deadlines

Organized by course with assignment types, dates, and descriptions.

Contacts

Professor names and email addresses for each course.

How It Works

1. User Input: Type a question in the text field
2. Query Processing: handle_query() function parses the message
3. Keyword Matching: Searches for relevant keywords (schedule, deadline, contact, course names)
4. Response Generation: Returns formatted information from data dictionaries
5. UI Update: Displays response in chat bubble format

Key Features Demonstrated

Flet UI Components

• ft.Column for scrollable chat history
• ft.Container for styled message bubbles
• ft.TextField for user input
• ft.ElevatedButton for send action

Natural Language Understanding

• Keyword mapping (e.g., "eap" -> "English for Academic Purposes")
• Case-insensitive query handling
• Multiple keyword variations support

User Experience

• Color-coded messages (blue for user, green for bot)
• Auto-scroll to latest message
• Clean, distraction-free interface

Project Structure

Chatbot-Flet/
├── chatbot.py          # Main documented application with UI and query logic
├── data.py             # Static data (schedules, deadlines, contacts)
├── dicts.py            # Alternative data structure
├── main.py             # Alternative earlier UI implementation
├── test_pytorch.py     # Experimental/test file
├── image.png           # Project image asset
├── screenshots/        # Application screenshots
│   ├── chatbot-welcome.png
│   ├── chatbot-schedule.png
│   ├── chatbot-deadlines.png
│   └── chatbot-contact.png
└── README.md           # This file

What I Learned

Python Flet Framework:

• Building desktop UIs with Python
• Event handling and state management
• Dynamic UI updates with page.update()

User Interface Design:

• Chat bubble layouts
• Scrollable content containers
• Input field and button interactions

Data Organization:

• Structuring nested dictionaries for complex data
• Keyword mapping for natural language understanding
• Query parsing and response formatting

Software Architecture:

• Separation of UI and logic
• Event-driven programming patterns
• Modular function design

Limitations & Future Improvements

Current Limitations:

• Data is hardcoded (not connected to university systems)
• No user authentication
• Limited to predefined queries
• English language only

Planned Enhancements:

• Connect to university API for real-time data
• Add user authentication and personalized schedules
• Implement AI-based natural language understanding (NLP)
• Multi-language support with translation
• Export chat history feature
• Push notifications for upcoming deadlines
• Mobile app version (Flutter)
• Database integration for persistent data
• Admin panel for updating information

Technical Challenges Solved

1. Dynamic UI Updates: Used Flet's page.update() to refresh chat display after each message
2. Keyword Ambiguity: Created mapping system to handle abbreviations and variations
3. Data Formatting: Structured responses to be readable in chat format
4. User Input Validation: Handled empty messages and unknown queries gracefully

Use Cases

• Students: Quick access to schedules without logging into multiple systems
• Study Groups: Coordinating meeting times based on free periods
• Deadline Management: Checking upcoming assignments across all courses
• Professor Communication: Finding contact details instantly

Academic Context

Course: Project Study (RESE-1160)
Institution: University of Greenwich
Year: 2023-2024
Purpose: Prototype for improving student information access

License

Academic project for educational purposes.

Contact

Francisco Navarro
Email: frannavarrogil@gmail.com
GitHub: @franng95
LinkedIn: franjng

---

Note: This is a prototype with sample data from Term 3, 2023. Data is not connected to live university systems. For actual course information, refer to official university portals.