Update README.md
Browse files
README.md
CHANGED
|
@@ -7,26 +7,69 @@ sdk: static
|
|
| 7 |
pinned: false
|
| 8 |
---
|
| 9 |
|
| 10 |
-
<
|
|
|
|
|
|
|
| 11 |
|
|
|
|
| 12 |
|
| 13 |
-
|
| 14 |
-
|
|
|
|
| 15 |
|
| 16 |
-
|
|
|
|
|
|
|
| 17 |
|
| 18 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
| 19 |
|
| 20 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 21 |
|
| 22 |
We summarize information with Topic Modeling & Generative AI for RAG.
|
| 23 |
-
This provides an overview of your dataset contents in the blink of an eye
|
| 24 |
|
| 25 |
-
<
|
|
|
|
|
|
|
| 26 |
|
| 27 |
-
|
| 28 |
|
| 29 |
We project information on a supervised Axis to Explore textual data in a completely new way.
|
| 30 |
-
This allows you to investigate potential biases, or filter the content of your dataset in order to clean it faster
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 31 |
|
| 32 |
-
|
|
|
|
| 7 |
pinned: false
|
| 8 |
---
|
| 9 |
|
| 10 |
+
<p align="center">
|
| 11 |
+
<img src="bunka_logo.png" alt="Bunka Logo" width="600"/>
|
| 12 |
+
</p>
|
| 13 |
|
| 14 |
+
<h1 align="center">Welcome to Bunka</h1>
|
| 15 |
|
| 16 |
+
<p align="center">
|
| 17 |
+
Bunka provides visual investigation of textual datasets, using Topic Modeling and Frame Analysis.
|
| 18 |
+
</p>
|
| 19 |
|
| 20 |
+
<p align="center">
|
| 21 |
+
Whether you want to understand your training datasets, its content or what it's missing before fine-tuning your model, you're at the right place!
|
| 22 |
+
</p>
|
| 23 |
|
| 24 |
+
<p align="center">
|
| 25 |
+
<a href="https://www.bunka.ai/">π Website</a> |
|
| 26 |
+
<a href="https://github.com/charlesdedampierre/BunkaTopics">π GitHub</a> |
|
| 27 |
+
<a href="https://beta.bunkasearch.com/">π Bunka Platform</a>
|
| 28 |
+
</p>
|
| 29 |
|
| 30 |
+
<p align="center">
|
| 31 |
+
<a href="https://github.com/charlesdedampierre/BunkaTopics"><img src="https://img.shields.io/badge/GitHub-100000?style=for-the-badge&logo=github&logoColor=white" alt="GitHub"></a>
|
| 32 |
+
<a href="https://www.linkedin.com/company/85881815"><img src="https://img.shields.io/badge/LinkedIn-0077B5?style=for-the-badge&logo=linkedin&logoColor=white" alt="LinkedIn"></a>
|
| 33 |
+
<a href="https://discord.gg/3YRPVqXabQ"><img src="https://img.shields.io/badge/Discord-7289DA?style=for-the-badge&logo=discord&logoColor=white" alt="Discord"></a>
|
| 34 |
+
<a href="https://huggingface.co/bunkalab"><img src="https://img.shields.io/badge/π€%20Hugging%20Face-ffce6b?style=for-the-badge" alt="Hugging Face"></a>
|
| 35 |
+
</p>
|
| 36 |
+
|
| 37 |
+
## π Explore Our Platform
|
| 38 |
+
|
| 39 |
+
<p align="center">
|
| 40 |
+
<a href="https://beta.bunkasearch.com/">
|
| 41 |
+
<img src="platform_hub.png" alt="Bunka Platform Overview" width="800"/>
|
| 42 |
+
</a>
|
| 43 |
+
</p>
|
| 44 |
+
|
| 45 |
+
## π§ Topic Modeling
|
| 46 |
|
| 47 |
We summarize information with Topic Modeling & Generative AI for RAG.
|
| 48 |
+
This provides an overview of your dataset contents in the blink of an eye!
|
| 49 |
|
| 50 |
+
<p align="center">
|
| 51 |
+
<img src="newsmap.png" alt="Topic Modeling Example" width="600"/>
|
| 52 |
+
</p>
|
| 53 |
|
| 54 |
+
## πΌοΈ Frame Analysis
|
| 55 |
|
| 56 |
We project information on a supervised Axis to Explore textual data in a completely new way.
|
| 57 |
+
This allows you to investigate potential biases, or filter the content of your dataset in order to clean it faster!
|
| 58 |
+
|
| 59 |
+
<p align="center">
|
| 60 |
+
<img src="bourdieu.png" alt="Frame Analysis Example" width="600"/>
|
| 61 |
+
</p>
|
| 62 |
+
|
| 63 |
+
## π¬ Example: IMDB Dataset Visualization
|
| 64 |
+
|
| 65 |
+
Explore our visualization of the IMDB dataset:
|
| 66 |
+
|
| 67 |
+
<p align="center">
|
| 68 |
+
<a href="https://beta.bunkasearch.com/map/206">
|
| 69 |
+
<img src="imdb_dataset.png" alt="IMDB Dataset Visualization" width="800"/>
|
| 70 |
+
</a>
|
| 71 |
+
</p>
|
| 72 |
+
|
| 73 |
+
## π€ Join Our Community
|
| 74 |
|
| 75 |
+
We're excited to have you join our community! Feel free to reach out on any of our platforms for questions, suggestions, or collaborations.
|