NeuroGuard FND — Fake News Detection
PyPI Package: fakenews-ussu321
Python Package: fakenews_ussu321
Version: 2.0.0
Developer: Mohammed Usman
GitHub: @issu321
Website: issu321.github.io
Overview
NeuroGuard FND is a professional desktop application for detecting fake news using an Enhanced Ensemble NLP model. It combines TF-IDF vectorization, statistical feature engineering, sentiment analysis, and machine learning to classify news articles as Real or Fake with confidence scores and explainable AI insights.
Features
- Single Article Analysis — Paste any news article and get instant real/fake classification with confidence score
- Batch Processing — Upload CSV or Excel files to analyze multiple articles at once
- Explainable AI — Get detailed explanations of why an article was flagged (sensationalist language, emotional tone, capitalization, etc.)
- Sentiment Analysis — Real-time polarity and subjectivity scoring via TextBlob
- User Authentication — Secure registration and login with Flask-Login
- Desktop App — Native desktop window powered by pywebview (no browser needed)
- Web Server Mode — Optional
--server-onlyflag for browser-based deployment - Model Statistics API — Compare ensemble model performance metrics
Installation
pip install fakenews-ussu321
Usage
Desktop Mode (Default)
fakenews
Or:
python -m fakenews_ussu321
Launches a native desktop window at http://127.0.0.1:5000/.
Web Server Mode
fakenews --server-only
Or:
python -m fakenews_ussu321 --server-only
Runs the Flask development server accessible from any browser.
Package Structure
fakenews_ussu321/
├── __init__.py # Package metadata
├── __main__.py # python -m entry point
├── cli.py # CLI entry point (fakenews command)
├── desktop.py # pywebview desktop launcher
├── app.py # Flask application factory
├── config.py # Configuration settings
├── models/
│ ├── __init__.py # SQLAlchemy & LoginManager init
│ ├── predictor.py # FakeNewsPredictor (ensemble NLP model)
│ └── user.py # User database model
├── routes/
│ ├── __init__.py # Blueprint registration
│ ├── main.py # Public pages (index, about, contact, features)
│ ├── auth.py # Authentication (login, register, logout)
│ └── dashboard.py # Prediction API & dashboard routes
├── templates/ # Jinja2 HTML templates
├── static/ # CSS, JS, images
├── Datasets/ # Sample datasets
├── instance/ # SQLite database storage
└── *.pkl # Trained ML models
Dependencies
| Package | Version |
|---|---|
| Flask | >=2.3.0 |
| Flask-SQLAlchemy | >=3.0.0 |
| Flask-Login | >=0.6.0 |
| Werkzeug | >=2.3.0 |
| pandas | >=2.0.0 |
| numpy | >=1.24.0 |
| scikit-learn | >=1.3.0 |
| joblib | >=1.3.0 |
| textblob | >=0.17 |
| openpyxl | >=3.1.0 |
| pywebview | >=4.4.0 |
Model Architecture
The Enhanced Ensemble NLP model extracts the following features:
- TF-IDF Vectorization — Term frequency-inverse document frequency
- Statistical Features (6 features):
- Capitalization ratio
- Exclamation mark count
- Question mark count
- Word count
- Average word length
- Sensationalist word count
- Sentiment Features (2 features):
- Polarity (via TextBlob)
- Subjectivity (via TextBlob)
All features are combined into a sparse matrix and fed into an ensemble classifier.
API Endpoints
| Endpoint | Method | Auth | Description |
|---|---|---|---|
/ |
GET | No | Landing page |
/features |
GET | No | Features page |
/about |
GET | No | About page |
/contact |
GET | No | Contact page |
/auth/register |
GET/POST | No | User registration |
/auth/login |
GET/POST | No | User login |
/auth/logout |
GET | Yes | User logout |
/auth/api/check-username/<username> |
GET | No | Username availability |
/dashboard/ |
GET | Yes | Main dashboard |
/dashboard/api/predict |
POST | Yes | Single article prediction |
/dashboard/api/predict-batch |
POST | Yes | Batch file prediction |
/dashboard/api/stats |
GET | Yes | Model performance stats |
/api/model-stats |
GET | No | Best model statistics |
Developer
Mohammed Usman
- Email: jaafreeusman@gmail.com
- Phone: 8884294749
- GitHub: issu321
- Website: issu321.github.io
License
MIT License
Related Packages
- algo-ussu321 — Algorithm Visualizer
- programmingvisualizer-ussu321 — Programming Visualizer
- aidataviz-ussu321 — AI Data Visualization
- business-ussu321 — Business Simulator
Metadata
Release files for fakenews-ussu321 2.0.0
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | Actions |
|---|---|---|---|---|
| fakenews_ussu321-2.0.0.tar.gz | 374.4 kB | Without provenance | Details for fakenews_ussu321-2.0.0.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| fakenews_ussu321-2.0.0-py3-none-any.whl | Python 3 | none | any | Details for fakenews_ussu321-2.0.0-py3-none-any.whl |
Total release size: 763.9 kB
Release files / fakenews_ussu321-2.0.0.tar.gz
| Download URL | fakenews_ussu321-2.0.0.tar.gz |
|---|---|
| Size | 374.4 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
bf0a49fec16496229d41d9413ec46eba1e88f2ef6448b59be62a9dc8c4525538
|
|
BLAKE2b-256 checksum How to use checksums |
e8b2f39cf2fc44a8d49ce8fda7c31c7bf586f09ac99e2d0d1e0770e410e0d6a4
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.2.0 CPython/3.13.14
|
Release files / fakenews_ussu321-2.0.0-py3-none-any.whl
| Download URL | fakenews_ussu321-2.0.0-py3-none-any.whl |
|---|---|
| Size | 389.6 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
286f6fa710821fe998fe8eee9251a90f188eac6519824deb99eb87e24a2e13eb
|
|
BLAKE2b-256 checksum How to use checksums |
eee9497e1c5a7642e6a5cb67f5cf2cd3d623ca9a322b46f7f88226965422a7e9
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.2.0 CPython/3.13.14
|
Security for fakenews-ussu321 2.0.0
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.