Update app.py

app.py

@@ -1,5 +1,6 @@
+import gradio as gr
 import os
-import fitz  # Corrected import statement
+import fitz
 import nltk
 from nltk.tokenize import word_tokenize
 from nltk.corpus import stopwords
@@ -14,9 +15,6 @@ nltk.download('stopwords')
 dataset_dir = '/content/LegalData'
 
 # Load and preprocess the query
-query = "What are the legal implications of intellectual property rights?"
-
-# Function to extract text from PDFs using PyMuPDF (fitz)
 def extract_text_from_pdf(pdf_path):
     pdf_text = ""
     with fitz.open(pdf_path) as pdf_document:
@@ -31,7 +29,7 @@ def clean_and_tokenize(text):
     tokens = [word for word in tokens if word.isalnum() and word not in stopwords.words('english')]
     return ' '.join(tokens)
 
-# Process and tokenize the documents in your dataset
+# Load and preprocess the documents in your dataset
 documents = []
 for filename in os.listdir(dataset_dir):
     if filename.endswith('.pdf'):
@@ -44,22 +42,37 @@ for filename in os.listdir(dataset_dir):
 tfidf_vectorizer = TfidfVectorizer()
 tfidf_matrix = tfidf_vectorizer.fit_transform(documents)
 
-# Vectorize the query
-query_vector = tfidf_vectorizer.transform([clean_and_tokenize(query)])
-
-# Calculate cosine similarities between the query and documents
-cosine_similarities = cosine_similarity(query_vector, tfidf_matrix)
-
-# Rank documents by similarity score
-document_scores = list(enumerate(cosine_similarities[0]))
-sorted_documents = sorted(document_scores, key=lambda x: x[1], reverse=True)
+# Function to perform relevance matching and return top N documents
+def perform_relevance_matching(query, uploaded_files):
+    # Vectorize the query
+    query_vector = tfidf_vectorizer.transform([clean_and_tokenize(query)])
+    
+    # Calculate cosine similarities between the query and documents
+    cosine_similarities = cosine_similarity(query_vector, tfidf_matrix)
+    
+    # Rank documents by similarity score
+    document_scores = list(enumerate(cosine_similarities[0]))
+    sorted_documents = sorted(document_scores, key=lambda x: x[1], reverse=True)
+    
+    # Extract the top N relevant documents
+    top_n = 5
+    top_documents = []
+    for i in range(min(top_n, len(sorted_documents))):
+        doc_index, score = sorted_documents[i]
+        document_text = documents[doc_index][:500]  # Extract the first 500 characters of the document
+        top_documents.append((f"Document {doc_index + 1} (Similarity Score: {score:.4f})", document_text))
+    
+    return top_documents
 
-# Print the top N relevant documents
-top_n = 5
-for i in range(top_n):
-    doc_index, score = sorted_documents[i]
-    print(f"Document {doc_index + 1} (Similarity Score: {score:.4f})")
-    print(documents[doc_index][:500])  # Print the first 500 characters of the document
-    print("\n")
+# Create a Gradio interface
+iface = gr.Interface(
+    fn=perform_relevance_matching,
+    inputs=["text", gr.File()],
+    outputs=gr.Table(),
+    live=True,
+    title="Legal Research Assistant",
+    description="Enter your legal query and upload files for relevance matching.",
+)
 
-# Implement answer extraction and answer generation steps for the top N documents.
+# Launch the Gradio interface
+iface.launch()