Hi, I'm Noor-ul-Ain Maryam.
I Design. I Build. I Learn.

import numpy as np
import pandas as pd
from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.metrics.pairwise import cosine_similarity
from sklearn.metrics import accuracy_score

# Step 1: Load data from files
original_file_path = r'c:\Users\user\Desktop\Khadija\original.txt'
plagiarized_file_path = r'c:\Users\user\Desktop\Khadija\plagiarized.txt'

with open(original_file_path, 'r', encoding='utf-8') as f:
    original_texts = f.readlines()

with open(plagiarized_file_path, 'r', encoding='utf-8') as f:
    plagiarized_texts = f.readlines()

if len(original_texts) != len(plagiarized_texts):
    raise ValueError("The two files must have the same number of documents.")

df = pd.DataFrame({'original': original_texts, 'plagiarized': plagiarized_texts})

# Step 2: Preprocess and vectorize text
vectorizer = TfidfVectorizer(stop_words='english', ngram_range=(1, 2))
original_vectors = vectorizer.fit_transform(df['original'])
plagiarized_vectors = vectorizer.transform(df['plagiarized'])

# Step 3: Compute similarity
similarity_scores = cosine_similarity(original_vectors, plagiarized_vectors)
df['similarity_score'] = np.diagonal(similarity_scores)
df['similarity_percentage'] = (df['similarity_score'] * 100).round(2)

# Step 4: Threshold-based plagiarism detection
threshold = 80.0
df['plagiarized_detected'] = df['similarity_percentage'] >= threshold

# Step 5: Evaluate accuracy
y_true = [True] * len(df)
y_pred = df['plagiarized_detected'].tolist()
accuracy = accuracy_score(y_true, y_pred) * 100

print("Plagiarism Detection Results:")
print(df[['original', 'plagiarized', 'similarity_percentage', 'plagiarized_detected']])
print(f"\nAccuracy: {accuracy:.2f}%")