Spaces:
Sleeping
Sleeping
File size: 1,644 Bytes
6b8b7e2 efa406c 6b8b7e2 |
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 |
---
title: Capstone Project
emoji: ⚡
colorFrom: gray
colorTo: green
sdk: gradio
sdk_version: 5.45.0
app_file: app.py
pinned: false
short_description: AI vs Human text classifier
---
# 🤖 AI vs Human Text Classifier (RoBERTa)
This project fine-tunes **RoBERTa** to classify text as either:
- 🧑 Human-Written
- 🤖 AI-Generated
It was developed as a **Capstone Project** to explore the power of transformer-based models in detecting AI-generated content.
---
## 📌 Project Overview
With the rapid rise of LLMs like GPT and other AI text generators, distinguishing between human-written and AI-generated text is becoming crucial in education, research, and online authenticity.
This project leverages **RoBERTa**, a transformer-based model, to build a binary text classifier.
---
## 🛠️ Features
- Fine-tuned **RoBERTa-base** model
- Binary classification: `Human (0)` vs `AI (1)`
- Deployed with **Gradio** for easy interaction
- Model hosted on **Hugging Face Model Hub**
---
## 📂 Dataset
The dataset used in training contains two columns:
- **Text** → the input text sample
- **Generated** → label (`0 = Human`, `1 = AI`)
---
## 🚀 Training
The model was fine-tuned on Google Colab using the Hugging Face `transformers` library.
**Steps:**
1. Load dataset (`Text`, `Generated`)
2. Preprocess using Hugging Face `AutoTokenizer`
3. Fine-tune RoBERTa with `Trainer` API
4. Evaluate using Accuracy, Precision, Recall, F1-score
---
## 📊 Results
Validation accuracy achieved: **~99%**
Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
|