add keyword extraction

app.py

@@ -2,6 +2,7 @@ import streamlit as st
 import PyPDF2
 from extractive_model import summarize_with_textrank
 from abstractive_model import summarize_with_bart
+from keyword_extraction import extract_keywords
 #from blanc import BlancHelp
 
 # Set page to wide mode
@@ -24,7 +25,7 @@ def main():
 
     # Left column: Radio buttons for summarizer choice
     with col1:
-        radio_options = ['Abstractive', 'Extractive']
+        radio_options = ['Abstractive', 'Extractive', 'Keyword Extraction']
         radio_selection = st.radio("Choose type of summarizer:", radio_options)
 
     # Middle column: Text input and File uploader
@@ -55,6 +56,11 @@ def main():
                 summary = summarize_with_bart(file_content)
                 st.session_state.summary = summary
 
+            # Perform extractive summarization
+            if radio_selection == "Keyword Extraction":
+                summary = extract_keywords(file_content)
+                st.session_state.summary = summary
+
     # Right column: Displaying text after pressing 'Summarize'
     with col3:
         st.write("Summary:")

keyword_extraction.py

@@ -0,0 +1,18 @@
+import nltk
+from rake_nltk import Rake
+
+# Download NLTK data (if not already downloaded)
+nltk.download('punkt')
+
+def extract_keywords(text):
+    # Initialize Rake with stopwords set to None (to keep all words)
+    rake = Rake()
+    
+    # Extract keywords from the input text
+    rake.extract_keywords_from_text(text)
+    
+    # Get the ranked keywords
+    ranked_keywords = rake.get_ranked_phrases()
+    
+    return ranked_keywords
+