🌍 Anticipatory Action AI Chatbot

The Anticipatory Action AI Chatbot is a Streamlit-based application designed to assist disaster managers in planning and decision-making for drought-related scenarios. This AI-powered chatbot leverages OpenAI's GPT model to provide tailored insights and recommendations based on uploaded documents and selected framework sections.

---

Features

• Framework-Based Guidance: Choose from predefined sections of the Anticipatory Action Framework for focused support.
• Document Upload: Upload relevant PDF documents (e.g., guidelines, strategies) to provide the chatbot with contextual information.
• OpenAI Integration: Uses OpenAI's GPT model for accurate and actionable advice.
• Interactive Chat: Input queries and receive responses in a conversational format.
• Session Management: Retains chat history during the session for easy reference.

---

How to Use

1. Prerequisites

• Ensure you have Python installed.
• Obtain an OpenAI API key from the OpenAI API Platform.

2. Installation

1. Clone this repository:

   git clone git@github.com:nitinmagima/AA_CoPilot.git
   cd anticipatory-action-chatbot

2. Install the required dependencies:

   pip install -r requirements.txt

3. Running the Application

1. Start the Streamlit app:

   streamlit run app.py

2. Open the app in your browser. The default address is http://localhost:8501.

4. Using the Chatbot

• Enter API Key: Add your OpenAI API key in the sidebar.
• Select a Framework Section: Use the radio buttons to choose the section you want to focus on.
• Upload Documents: Upload relevant PDFs to provide additional context for the chatbot.
• Ask Questions: Use the input box to interact with the chatbot. Example queries include:
  • "What steps are required to activate anticipatory actions?"
  • "How can we use forecasting models in drought management?"

---

Application Flow

1. Configure OpenAI API Key:

   • The API key is required to access OpenAI's GPT model.

2. Select Framework Section:

   • Choose from key sections like:
     • Defining Risk and Establishing a Crisis Timeline.
     • Development of Anticipatory Actions Interventions.
     • Activating the Monitoring System of Anticipatory Actions.

3. Upload Documents:

   • Extracts text from uploaded PDFs for tailored advice based on the document content.

4. Chat and Get Insights:

   • Combines the selected section, document content, and user queries to generate responses.

---

Code Highlights

Key Components

1. Sidebar:

   • Collects the OpenAI API key.
   • Provides radio buttons for selecting the framework section.

2. File Upload:

   • Allows users to upload PDF documents. Extracted content is used to enhance the chatbot's context.

3. Chat Interaction:

   • Maintains session state to store chat history.
   • Sends user queries and document context to OpenAI for generating responses.

4. OpenAI Integration:

   • Combines user input, selected framework section, and document content into a comprehensive prompt for OpenAI's GPT model.

---

Example Queries

• "What are the key steps to activate anticipatory actions for drought forecasts?"
• "How can we integrate forecasting models into decision-making processes?"
• "What are the best practices for monitoring and evaluation in drought anticipatory actions?"

---

Troubleshooting

• Missing API Key:

  • If the API key is not provided, the app will prompt you to enter one in the sidebar.

• File Upload Issues:

  • Ensure the uploaded document is in PDF format and does not exceed the size limit.

• Response Delays:

  • Responses depend on OpenAI's API. Check your API usage limits or internet connection if delays occur.

---

License

This project is licensed under the MIT License.