Spaces:
Running
Running
File size: 1,848 Bytes
8c68bc5 5224131 8c68bc5 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 |
---
license: mit
title: Social Aegis
sdk: gradio
emoji: ⚡
colorFrom: blue
colorTo: yellow
pinned: true
sdk_version: 5.44.1
---
# SocialAegis MVP
🛡️ A sentiment-based escalation engine to detect emotional volatility in social media posts.
---
## Overview
SocialAegis is designed to analyze text from social media and other sources to detect sentiment, calculate a risk score, and determine whether escalation is needed. This helps organizations monitor emotional volatility in communications and take timely action.
---
## Features
- **Sentiment Analysis** – Classifies input text as Positive, Neutral, or Negative.
- **Risk Scoring** – Provides a numerical score indicating the emotional intensity or risk level.
- **Escalation Trigger** – Indicates whether the text requires escalation based on the risk score.
- **Supports All Text Inputs** – Works with short posts, long messages, and multiple languages.
---
## How to Use
1. Enter your text into the input box.
2. Click **Submit** or **Run**.
3. The app will return:
- **Sentiment** (Positive / Neutral / Negative)
- **Risk Score** (rounded to 2 decimal places)
- **Escalation Trigger** (Yes / No)
### Example Inputs
- `"I am so frustrated with this service!"`
- `"I had an amazing experience, thank you!"`
- `"This is unacceptable, I will report this."`
---
## Dependencies
The app requires the following Python packages (all included in `requirements.txt`):
- `gradio`
- `transformers`
- `torch`
- `scipy`
- `sentencepiece`
- `protobuf`
- `langdetect`
---
## Notes
- The model runs entirely in Hugging Face Spaces. No additional setup is required if you use the provided `requirements.txt`.
- For best results, enter text in a single language at a time.
---
## Credits
Developed by Karthik Vinay for SocialAegis MVP. |