Authors: Aman Narendra Ali and Vir Khanna
- About
- Features
- Installation & Setup
- Usage
- FAQs
- Contributions
- License
- Contact
ChatCKB is a unique twist on OpenAI's advanced AI model, ChatGPT. Specifically tailored for CK Birla, it has the ability to comprehend and process company documents and answer relevant questions based on this acquired knowledge. The AI is powered by the OpenAI API and LangChain.
-
Document Analysis: ChatCKB can efficiently parse through a vast amount of company documentation and quickly gain valuable insights.
-
Intelligent Response: Leveraging the power of GPT-4, the chatbot can answer intricate queries, providing accurate and insightful responses in a natural, human-like manner.
-
24/7 Availability: Unlike human counterparts, ChatCKB is available round-the-clock, capable of answering queries at any time of the day.
-
Easy Integration: The chatbot can be seamlessly integrated into various platforms, such as websites, apps, or customer service interfaces, providing an enriching user experience.
Ensure you have Python installed (Python 3.6 or newer is recommended). You'll also need the following Python packages which can be installed via pip:
- os
- PyPDF2
- langchain
- openai
Clone the repository: git clone https://github.com/AmanNarendraAli/ChatCKB.git
Navigate into the cloned repository: cd ChatCKB
Install the necessary packages using pip: pip install -r requirements.txt
Set up your OpenAI API key as an environment variable in your operating system.
For Unix-based systems, you can add the following line to your .bashrc or .bash_profile:
export OPENAI_API_KEY='your-api-key'
Replace 'your-api-key' with your actual OpenAI API key.
ChatCKB provides a conversational interface to interact with the user. Upon execution, it will prompt you to enter your question.
Here are some examples:
"What is the company's mission as stated in document X?"
"Can you summarise the main points from document Y?"
Remember to be clear and specific with your questions to get the most accurate responses.
For any additional questions, please refer to the Contact section.
We welcome contributions from everyone. To maintain code consistency, we request you to follow these guidelines:
Use clear, descriptive commit messages. Refer to our commit messages for examples of what your clear, descriptive commit messages should look like.
Follow the PEP8 style guide for Python code.
Include comments and docstrings for functions and classes.
To propose new features or improvements, please create a new issue on GitHub detailing your proposal.
ChatCKB is currently open-source and free for non-commercial use. For commercial use, please contact the authors.
For further information or queries regarding ChatCKB, feel free to contact us via the following methods:
Aman Narendra Ali: [email protected]
Vir Khanna: [email protected]
Aman Ali: AmanNarendraAli
Vir Khanna: v-khanna
Raise an issue on the repository
Aman Narendra Ali: @amannnali
Vir Khanna: @virkhanna
Vir Khanna: @vir_khanna