https://github.com/besser-pearl/databot
This platform is used to create bots whose job is to answer questions about a specific data source. It allows the automatic generation of a chat/voice bot swarm to attend all the data sources in an Open Data Portal.
https://github.com/besser-pearl/databot
chatbot datasets intent-detection llm opendata voicebot
Last synced: 3 months ago
JSON representation
This platform is used to create bots whose job is to answer questions about a specific data source. It allows the automatic generation of a chat/voice bot swarm to attend all the data sources in an Open Data Portal.
- Host: GitHub
- URL: https://github.com/besser-pearl/databot
- Owner: BESSER-PEARL
- License: mit
- Created: 2023-10-27T19:39:59.000Z (over 2 years ago)
- Default Branch: main
- Last Pushed: 2024-11-18T17:47:00.000Z (over 1 year ago)
- Last Synced: 2026-04-13T07:48:56.762Z (3 months ago)
- Topics: chatbot, datasets, intent-detection, llm, opendata, voicebot
- Language: Python
- Homepage:
- Size: 245 KB
- Stars: 8
- Watchers: 1
- Forks: 3
- Open Issues: 1
-
Metadata Files:
- Readme: README.md
- Contributing: CONTRIBUTING.md
- Funding: FUNDING.yml
- License: LICENSE.md
- Code of conduct: CODE_OF_CONDUCT.md
- Citation: CITATION.cff
- Codeowners: CODEOWNERS
- Security: SECURITY.md
- Governance: GOVERNANCE.md
Awesome Lists containing this project
README
# DataBot: Reliable data exploration through chatbots
This platform is used to create bots whose job is to answer questions about a specific data source. It allows the automatic
generation of a chat/voice bot swarm to attend all the data sources in an **Open Data Portal**.
The highlights of DataBot are:
- 💻 **Import data** through a friendly UI.
- 💾 Upload your dataset directly to the platform, or...
- 🌐 Automatically load all the data sources from an Open Data Portal through its API.
- 🔎 A **data schema** is automatically inferred from the data source, and can be **enhanced** 💪 to improve the bot knowledge about
the data (e.g., synonyms or translations). This can be done either manually or using ✨AI.
- 🤖 **Automatically generate a chatbot for each data source**. These chatbots are powered by the [**BESSER Bot Framework**](https://github.com/BESSER-PEARL/bot-framework).
They recognize the user intent and generate the appropriate answer. So, no hallucinations at all.
- Generation of tabular📅 and graphical📈 answers.
- 🎙️ Interact with the chatbots either writing or speaking: **voice recognition integrated**.
- ✨ For those questions the bot fails to identify, AI can be used to generate the best possible answer.
- ✨ For the AI components (data schema enhancement and answer generation), we use the OpenAI API.
### Requirements
- Python 3.11
- Recommended: Create a virtual environment (e.g. [venv](https://docs.python.org/3/library/venv.html), [conda](https://conda.io/projects/conda/en/latest/user-guide/tasks/manage-environments.html))
### Installation
```bash
git clone https://github.com/BESSER-PEARL/databot
cd databot
pip install -r requirements.txt
python main.py
```
### License
This project is licensed under the [MIT](https://mit-license.org/) license
Copyright © 2023 Luxembourg Institute of Science and Technology. All rights reserved.