PROJECT GOALS
Our Project Focuses On Advancing Urdu Back Translation Text Reuse Detection By Creating A Specialized Back-Translated Dataset. This Dataset Addresses The Unique Challenges Of Detecting Text Reuse In The Urdu Language. The Key Objectives Of Our Project Are:
MEET PROJECT TEAM
DR
MUHAMMAD SHARJEEL
Assistant Professor at COMSATS University Islamabad, Lahore Campus, specializing in Natural Language Processing, Urdu NLP, and Text reuse Detection. Focused on advancing machine learning solutions for low-resource languages.
Project Documentation
Access our comprehensive project documentation and reports
Project Proposal
2024-09-15
Initial project proposal document outlining the scope and objectives
Download DocumentHigh Fidelity Testing Report
2024-02-10
Comprehensive testing documentation and results
Download DocumentOpen Source Research
Explore our complete codebase and experimental results on GitHub. We believe in open science and sharing our research with the community.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
# Urdu Back Translation Text Reuse Detection
from transformers import AutoTokenizer, AutoModel
import torch
def analyze_text_similarity(text1, text2):
# Load pretrained model for Urdu
model_name = "sentence-transformers/LaBSE"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModel.from_pretrained(model_name)
# Process texts
encoded = tokenizer([text1, text2],
padding=True,
truncation=True,
return_tensors="pt")OUR TECH STACK
Next.js
Frontend Framework
React
UI Library
Tailwind CSS
Styling Framework
Python
Core Processing
Machine Learning
ML Experiments
MongoDB
Database