Apppublic

theyashkhatri/AuthenticPlagAI

sourceHugging Faceupdated 8mo agoView on Hugging Face

0likes

App README

🛡️ AuthenticPlagAI

An AI-Driven Plagiarism Detection and Content Authentication System with Humanization Module

A clean, single-file Streamlit app to detect plagiarism in any text or document and rewrite AI-generated content to sound naturally human.

🚀 Quick Start

bash

# 1. Install dependencies
pip install -r requirements.txt

# 2. Run the app
streamlit run app.py

Opens at http://localhost:8501

✨ Features

📝 Check Text

Paste any essay, article, or content directly into the text box and click Check Plagiarism to instantly get a plagiarism score with a full breakdown.

📂 Check File

Upload a .txt, .pdf, or .docx file. A live preview of the document appears before you run the check.

✨ AI Humaniser

Paste AI-generated or flagged text and rewrite it to sound authentically human using Google Gemini.

—5 tone options: Natural, Academic, Professional, Creative, Plain
—3 intensity levels: Light Touch, Moderate, Heavy Rewrite
—Side-by-side original vs humanised view
—Download the result as .txt

📊 How Plagiarism Scoring Works

The score is computed using three internal signals:

Signal	Weight	Description
N-Gram Repetition	50%	How much of the text repeats itself in 5-word chunks
Stopword Ratio	25%	Proportion of generic/common words
Sentence Uniformity	25%	How uniform sentence lengths are (AI text tends to be very uniform)

Score	Verdict
< 25%	✅ Likely Original
25–55%	⚠️ Moderate Risk
> 55%	🚨 High Plagiarism Risk

🛠️ Tech Stack

Layer	Technology
UI	Streamlit
Plagiarism Analysis	scikit-learn, Python difflib
AI Humaniser	Google Gemini 2.0 Flash
PDF Parsing	pdfplumber
DOCX Parsing	python-docx

🔑 Gemini API Key (Humaniser only)

The plagiarism checker works without any API key. The Humaniser tab requires a free Google Gemini API key:

1.Go to aistudio.google.com
2.Click Get API Key → Create API Key
3.Paste it into the sidebar of the app

Free tier: 15 requests/min, 1M tokens/day — no credit card needed.

📦 Requirements

streamlit>=1.32.0
scikit-learn>=1.4.0
pandas>=2.0.0
google-generativeai>=0.7.0
pdfplumber>=0.10.0
python-docx>=1.1.0

☁️ Deploy to Hugging Face Spaces

1.Create a new Space at huggingface.co/spaces
2.Select Streamlit as the SDK
3.Upload app.py, requirements.txt, and README.md
4.The Space builds and deploys automatically

📂 Project Structure

AuthenticPlagAI/
├── app.py            # Entire application (single file)
├── requirements.txt  # Python dependencies
└── README.md         # This file

📄 License

MIT License — free to use, modify, and distribute.