🤖 RAG Chatbot — Generation Augmented Retrieval

---

🧠 Description

This project provides a Interactive chatbot based on RAG (Retrieval-Augmented Generation), designed to enhance question-answering tasks by retrieving and leveraging information from uploaded PDF documents.

Unlike traditional chatbots that rely solely on pre-trained language models, this approach combines retrieval and generation, enabling the model to produce more accurate, contextualized, and reliable responses based on real source material.

General flow:

1. Process and vectorize the PDF content. The PDF content is processed and transformed into embeddings using advanced language models, converting textual information into a numerical representation that captures semantic meaning.
2. Query the vector store to retrieve relevant fragments. When a user submits a query, the system searches the vector store to identify and retrieve the most relevant text fragments from the document.
3. Answer Generation. The retrieved fragments are passed to a language model (gemma:2b), which uses this contextual information to generate a coherent and well-informed response.

---

🏗️ System architecture

PDF --> Preprocessing --> Embeddings (Ollama) --> Vector Store (Chroma) User --> Recovery --> Model Gemma:2b --> Response contextualized

---

✨ Characteristics

• 📚 Support for multiple PDF files.
• 💬 Natural language query with rich context.
• 🧠 Using Ollama embeddings.
• 🤖 Generation with model gemma:2b.
• 🖥️ Simple web interface with Streamlit.
• 🧪 Optional integration with LangSmith.

---

🧰 Requirements

• Python 3.8 or superior
• Git
• Virtual environment (recommended)

---

⚙️ Installation

# 1. Clone the repository
git clone https://github.com/seba39399/Chatbot-RAG---Juan-Sebastian-Pena-V.git
cd Chatbot-RAG---Juan-Sebastian-Pena-V

# 2. Create virtual environment
python -m venv .venv

# 3. Activate virtual environment
# Windows
.\.venv\Scripts\activate
# Linux/macOS
source .venv/bin/activate

# 4. Install dependencies
pip install -r requirements.txt

## 🔐 Configuration (optional)

If you want to enable traceability with LangSmith:

```bash
# Windows
setx LANGCHAIN_API_KEY "tu_api_key_aqui"

# Linux/macOS
export LANGCHAIN_API_KEY="tu_api_key_aqui"

💬 Examples of use

Question: Who is the main character? Answer: The main character is Little Red Riding Hood.

Question: What is the story about? Answer: It's about how Little Red Riding Hood learns not to trust strangers and to follow her mother's instructions.

Question: How does the story end? Answer: It ends when Little Red Riding Hood arrives at her grandmother's house and encounters a wolf.

---

🤝 Contributions and credits

👨‍💻 Developed by Biomedical Engineer and Artificial Intelligence Specialist: Juan Sebastián Peña Valderrama

🚀 Inspired by RAG paradigms and open-source tools:

-LangChain

-Ollama

-Chroma

-Streamlit

🧩 Open contributions: issues, suggestions and PRs are welcome