Upload 2 files

app.py

@@ -0,0 +1,112 @@
+import streamlit as st
+import numpy as np
+import pandas as pd
+import seaborn as sns
+import matplotlib.pyplot as plt
+from sklearn.feature_extraction.text import TfidfVectorizer
+from sklearn.metrics.pairwise import cosine_similarity
+from sentence_transformers import SentenceTransformer
+import string
+import PyPDF2
+import docx
+
+# App title with improved styling
+st.set_page_config(page_title="Unsupervised Text Similarity Analysis", layout="wide")
+st.title("📖 Unsupervised Text Similarity Analysis")
+st.markdown("### Compare and analyze text similarity effortlessly! 🔍")
+
+# 🎯 Streamlit Tabs
+tab1, tab2 = st.tabs(["📖 About", "📊 Similarity Analysis"])
+
+# Load Sentence Transformer model
+@st.cache_resource
+def load_model():
+    return SentenceTransformer('all-MiniLM-L6-v2')
+
+model = load_model()
+
+def preprocess_text(text):
+    text = text.lower()
+    text = text.translate(str.maketrans('', '', string.punctuation))
+    return text
+
+def extract_text_from_pdf(file):
+    reader = PyPDF2.PdfReader(file)
+    text = " ".join([page.extract_text() for page in reader.pages if page.extract_text()])
+    return text
+
+def extract_text_from_docx(file):
+    doc = docx.Document(file)
+    text = " ".join([para.text for para in doc.paragraphs])
+    return text
+
+def compute_tfidf_similarity(texts):
+    vectorizer = TfidfVectorizer()
+    tfidf_matrix = vectorizer.fit_transform(texts)
+    return cosine_similarity(tfidf_matrix)
+
+def compute_bert_similarity(texts):
+    embeddings = model.encode(texts, convert_to_tensor=True)
+    return cosine_similarity(embeddings.cpu().numpy())
+
+def plot_similarity_matrix(similarity_matrix, labels):
+    fig, ax = plt.subplots(figsize=(10, 8))
+    sns.heatmap(similarity_matrix, annot=True, fmt='.2f', xticklabels=labels, yticklabels=labels, cmap='coolwarm')
+    plt.title("🔗 Text Similarity Matrix", fontsize=14)
+    st.pyplot(fig)
+
+# About Tab
+with tab1:
+    st.write("""
+    Welcome to the **Unsupervised Text Similarity Analysis** app! 🚀
+    This app allows you to compare the similarity between multiple text documents.
+    
+    ### How It Works:
+    1. **Upload text documents** (TXT, PDF, DOCX).
+    2. **Choose a similarity method** (TF-IDF or BERT Embeddings).
+    3. **Compute similarity** to generate a similarity matrix.
+    4. **Visualize results** with a heatmap and similarity percentages.
+    
+    📌 **Use Cases:** Plagiarism detection, document comparison, research analysis, and more!
+    """)
+
+# Similarity Analysis Tab
+with tab2:
+    st.subheader("📂 Upload Text Documents")
+    uploaded_files = st.file_uploader("Upload text documents", type=["txt", "pdf", "docx"], accept_multiple_files=True)
+    
+    if uploaded_files:
+        documents = []
+        doc_names = []
+        
+        for file in uploaded_files:
+            if file.type == "text/plain":
+                text = preprocess_text(file.read().decode("utf-8"))
+            elif file.type == "application/pdf":
+                text = preprocess_text(extract_text_from_pdf(file))
+            elif file.type == "application/vnd.openxmlformats-officedocument.wordprocessingml.document":
+                text = preprocess_text(extract_text_from_docx(file))
+            else:
+                continue
+            
+            documents.append(text)
+            doc_names.append(file.name)
+        
+        similarity_method = st.selectbox("🔎 Choose Similarity Method", ["TF-IDF", "BERT Embeddings"], index=0)
+        
+        if st.button("🚀 Compute Similarity"):
+            if similarity_method == "TF-IDF":
+                similarity_matrix = compute_tfidf_similarity(documents)
+            else:
+                similarity_matrix = compute_bert_similarity(documents)
+            
+            st.subheader("📊 Similarity Matrix")
+            plot_similarity_matrix(similarity_matrix, doc_names)
+            
+            st.subheader("🔗 Document Similarity Scores")
+            for i in range(len(documents)):
+                for j in range(i + 1, len(documents)):
+                    similarity_percentage = similarity_matrix[i, j] * 100
+                    st.write(f"✅ **{doc_names[i]}** and **{doc_names[j]}** have a similarity of **{similarity_percentage:.2f}%**")
+    else:
+        st.info("📌 Please upload at least two text documents to start the analysis.")

requirements.txt

@@ -0,0 +1,7 @@
+streamlit
+numpy
+pandas
+seaborn
+matplotlib
+scikit-learn
+sentence-transformers