https://github.com/zain-ul-din/urdu_pleg_checker

The Model is an advanced AI solution that efficiently detects and eliminates plagiarism in Urdu text.
https://github.com/zain-ul-din/urdu_pleg_checker

Last synced: 5 months ago
JSON representation

The Model is an advanced AI solution that efficiently detects and eliminates plagiarism in Urdu text.

• Host: GitHub
• URL: https://github.com/zain-ul-din/urdu_pleg_checker
• Owner: Zain-ul-din
• Created: 2023-07-11T20:36:48.000Z (over 2 years ago)
• Default Branch: master
• Last Pushed: 2023-07-17T10:44:06.000Z (over 2 years ago)
• Last Synced: 2025-07-05T15:07:07.395Z (5 months ago)
• Language: Jupyter Notebook
• Size: 5.22 MB
• Stars: 2
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: readme.md

Awesome Lists containing this project

README

          
# Urdu Plagiarism Checker

The UrduPlagCheckAI Model makes a significant contribution to the Urdu Adab language. It is an advanced AI solution that effectively identifies and eliminates plagiarism in Urdu text.

### Usage

```bash

> Copy `LSTM.ipynb` to `google colabs`.

> Run All cells

> download model and use it anywhere.

```

## Examples

*Text reuse is the act of composing new text from a previously published text*

![image](https://github.com/Zain-ul-din/Urdu_Pleg_Checker/assets/78583049/a6d28901-44c7-4dc1-81a3-507a9d4c6a44)

![image](https://github.com/Zain-ul-din/Urdu_Pleg_Checker/assets/78583049/2098c8ad-808a-40b0-8a83-748d8e9a14b0)

## Model Training Results

**Model Accuracy**

![Accuracy](https://github.com/Zain-ul-din/Urdu_Pleg_Checker/assets/78583049/8f0aedc2-fcfc-41b2-901a-b9136036cc58)

**Lose**

![Lose](https://github.com/Zain-ul-din/Urdu_Pleg_Checker/assets/78583049/e47937ff-3205-406e-b015-ccf57c586d19)

## Tools

The tools making this model work:

- `urduhack`: Python library for Urdu text preprocessing and normalization.

- `nlu`: Python library for natural language understanding.

- `gensim`: Python library for topic modeling and document similarity.

- `Keras`: Deep learning library for neural network models.

- `TensorFlow`: An open-source deep learning framework for building and training neural networks.

- `scikit-learn`: Python library for machine learning.

- `matplotlib` and `seaborn`: Python libraries for data visualization.

- `pandas` and `numpy`: Python libraries for data manipulation and computation.

- `Word2Vec`: Algorithm for generating word embeddings.

- `Tokenizer` and `pad_sequences`: Functions for text tokenization and padding in Keras.

- `EarlyStopping` and `ModelCheckpoint`: Keras callbacks for training control.

- `TensorBoard`: Visualization tool for monitoring model training in TensorFlow.

Make sure to provide appropriate links or references to give credit to the respective tools and their developers.

## Internal Architecture

![image](https://github.com/Zain-ul-din/Urdu_Pleg_Checker/assets/78583049/03787815-e42a-4c0e-b6f8-c2e49badda54)

[Want Help?](https://github.com/Zain-ul-din/Urdu_Pleg_Checker/issues)