https://github.com/yusufahmed9/bob-bot

Your Educational helper
https://github.com/yusufahmed9/bob-bot

artificial-intelligence llms nlp prompt-engineering python telegram

Last synced: 2 months ago
JSON representation

Your Educational helper

• Host: GitHub
• URL: https://github.com/yusufahmed9/bob-bot
• Owner: YusufAhmed9
• Created: 2025-07-26T12:22:25.000Z (11 months ago)
• Default Branch: main
• Last Pushed: 2025-07-26T17:39:15.000Z (11 months ago)
• Last Synced: 2025-07-26T19:35:05.846Z (11 months ago)
• Topics: artificial-intelligence, llms, nlp, prompt-engineering, python, telegram
• Language: Jupyter Notebook
• Homepage:
• Size: 19.5 MB
• Stars: 0
• Watchers: 0
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

          
# 🤖 Project Report: BoB Bot – Your AI-Powered Learning Assistant

## 📌 Project Title

**BoB Bot – AI Telegram Bot for Summarizing, Quizzing, and Finding Learning Resources**

---

## 📚 Objective

BoB Bot is a Telegram-based educational assistant designed to:

- 📑 Summarize lectures, PDFs, or custom topics

- ❓ Generate MCQs for revision and testing

- 🔎 Discover online resources (videos, notes, papers)

Built to support active learning with the help of local LLM inference (OpenHermes) and responsive user interaction.

---

## 🛠️ Tools & Technologies Used

| Tool / Tech              | Purpose                                |

|--------------------------|----------------------------------------|

| **Python**               | Core development language              |

| **TeleBot**              | Telegram bot framework                 |

| **OpenHermes via Ollama**| Local LLM for summarization & MCQs     |

| **LangChain**            | Prompt management, chat memory         |

| **PyMuPDF (fitz)**       | PDF text extraction                    |

| **Regular Expressions**  | Parsing answers & cleaning GPT output  |

| **YouTube API** (or scraping) | Finding relevant video content |

| **Google Search / SerpAPI** | Web-based learning material search  |

---

## ⚙️ Key Features

### 📄 1. Topic or PDF Input

- User can either **enter a topic** or **upload a PDF**.

- PDF contents are extracted and cleaned before processing.

### 🧠 2. Summarization

- The **OpenHermes** model generates a structured, clean summary of the content.

- Summaries are tailored to be short, informative, and beginner-friendly.

### ❓ 3. MCQ Generation and Quiz Mode

- BoB uses OpenHermes to generate **4-option MCQs** from any topic or document.

- Telegram inline buttons allow **interactive quiz mode** with:

  - Per-question feedback

  - Final evaluation/score after all answers

- Answers are validated with regex to ensure correctness extraction.

### 🌍 4. External Resource Discovery

- BoB finds **top relevant resources** for the user’s topic:

  - 📺 YouTube tutorials

  - 📖 Blog posts or academic notes

  - 📄 Research articles or books

- Each resource includes a title, short description, and link.