This repository contains two AI-powered models:
- SEO Rank Classifier (VariRank): Predict whether a web page is SEO-friendly, its probability, and expected rank (
top1,top3,top10,below10). - Viral Trend Detector: Detect whether a social media post is likely to go viral.
- Input HTML content in JSON format.
- Predict SEO friendliness (
true/false). - Get probability (%) and expected rank.
- Deployment via Hugging Face Spaces.
- Input post text in JSON format.
- Predict viral potential (
true/false). - Get probability (%) and predicted platform (Twitter, Facebook, etc.).
- Fast inference using ONNX models.
- Clone the repository:
git clone https://github.com/omar-abo-el-maaty/VariRank-AI-Models.git- Navigate to the project folder:
cd VariRank-AI-Models- Install dependencies:
pip install -r requirements.txt- Preprocess HTML content:
python scripts/seo_preprocess.py- Run predictions:
python scripts/app.py- Input JSON example:
{
"html": "<html><head><title>My Page</title></head><body>...</body></html>"
}- Example output:
{
"seo_friendly": true,
"probability": 92.3,
"expected_rank": "top3"
}Try online: Hugging Face App
- Preprocess text data:
python scripts/viral_trend_detector_v3.py- Input JSON example:
{
"text": "Check out this amazing AI project that will go viral!"
}- Example output:
{
"is_viral": true,
"probability": 87.5,
"platform": "Twitter"
}Try online: Hugging Face Viral Trend Detector
VariRank-AI-Models/
│
├─ model/
│ ├─ best_classification_model.onnx # SEO model
│ ├─ viral_trend_model.onnx # Viral trend model
│
├─ scripts/
│ ├─ seo_train_onnx_v11.py # SEO training script
│ ├─ seo_preprocess.py # SEO preprocessing
│ ├─ app.py # SEO main app
│ ├─ V_app.py # Viral trend detector app
│ └─ viral_trend_detector_v3.py # Viral trend detector
│
├─ data/ # Dataset for viral model
├─ tfidf.pkl # TF-IDF for viral model
├─ le_platform.pkl # Label encoder for viral model
├─ requirements.txt
└─ README.md
This project is licensed under the MIT License.