Update README.md
Browse files
README.md
CHANGED
@@ -1,6 +1,46 @@
|
|
1 |
---
|
|
|
2 |
license: apache-2.0
|
3 |
-
|
4 |
-
|
|
|
|
|
|
|
|
|
|
|
5 |
pipeline_tag: text-classification
|
6 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
---
|
2 |
+
language: en
|
3 |
license: apache-2.0
|
4 |
+
base_model: distilbert-base-uncased
|
5 |
+
tags:
|
6 |
+
- text-classification
|
7 |
+
- urgency-detection
|
8 |
+
- startup-grievances
|
9 |
+
- transformers
|
10 |
+
- huggingface
|
11 |
pipeline_tag: text-classification
|
12 |
+
widget:
|
13 |
+
- text: "Our server is down! We need immediate help!"
|
14 |
+
---
|
15 |
+
|
16 |
+
# 🚨 Urgency Model Aura (KS-Vijay)
|
17 |
+
|
18 |
+
This is a **DistilBERT-based text classification model** trained to detect the **urgency level** of textual grievances submitted by startups. It's part of an AI-based **Grievance Redressal Platform** for startups.
|
19 |
+
|
20 |
+
## 🧠 Use Case
|
21 |
+
|
22 |
+
Helps categorize complaints into urgency levels:
|
23 |
+
- `Low`
|
24 |
+
- `Medium`
|
25 |
+
- `High`
|
26 |
+
- `Critical`
|
27 |
+
|
28 |
+
This allows startups or organizations to **prioritize tickets** and **respond efficiently** based on severity.
|
29 |
+
|
30 |
+
## 🔍 How It Works
|
31 |
+
|
32 |
+
The model takes in a complaint text and outputs a classification label. It was trained on labeled grievance data using the 🤗 `transformers` and `datasets` libraries.
|
33 |
+
|
34 |
+
## 📦 Model Details
|
35 |
+
|
36 |
+
- Architecture: `DistilBERT`
|
37 |
+
- Framework: `PyTorch`
|
38 |
+
- Model type: `Text Classification`
|
39 |
+
- Format: `safetensors`
|
40 |
+
- Dataset: Custom `complaints.csv` (internal)
|
41 |
+
- Labels: Urgency levels (Critical, High, Medium, Low)
|
42 |
+
|
43 |
+
## 🚀 Example Input
|
44 |
+
|
45 |
+
```text
|
46 |
+
"My startup's payment system has been offline since morning!"
|