| --- |
| title: Individual vs Institution Classifier |
| emoji: π’ |
| colorFrom: blue |
| colorTo: green |
| sdk: streamlit |
| sdk_version: 1.32.0 |
| app_file: app.py |
| pinned: false |
| license: mit |
| --- |
| |
| # Individual vs Institution Classifier β JD Mart |
|
|
| Upload an Excel file with business names β download it back tagged with |
| **Individual** or **Institution** per listing. |
|
|
| ## How to use |
| 1. Upload your `.xlsx` file |
| 2. Click **Classify All** |
| 3. Download the tagged Excel with color-coded results |
|
|
| ## Output |
| - π’ Green = Individual (business depends on one person) |
| - π΅ Blue = Institution (multi-person, independent brand) |
| - π‘ Amber = Needs Review (low confidence) |
|
|
| ## Model |
| - Word TF-IDF + Character TF-IDF + Engineered Name Features |
| - Hybrid: Rule layer + Calibrated Logistic Regression |
| - Categories: Beauty Services, Healthcare, Home Services, Legal Services |
|
|