Update app.py

app.py

@@ -6,10 +6,15 @@ from langchain_core.output_parsers import StrOutputParser
 from langchain.chat_models import ChatOpenAI
 from fpdf import FPDF
 import os
+import spacy
+from collections import Counter
+
+# Load SpaCy model for NLP tasks (install 'spacy' and 'en_core_web_sm' first)
+nlp = spacy.load("en_core_web_sm")
 
 # Set up Streamlit UI
 st.title('Educational Assistant')
-st.header('Summary, Quiz Generator, Q&A, and Study Plan')
+st.header('Summary, Quiz Generator, Q&A, Study Plan, Vocabulary Builder & Topic Extraction')
 st.sidebar.title('Drop your PDF here')
 
 # Input OpenAI API key from keyboard
@@ -17,8 +22,8 @@ openai_api_key = st.sidebar.text_input("Enter your OpenAI API Key", type="passwo
 
 user_file_upload = st.sidebar.file_uploader(label='', type='pdf')
 
-# Sidebar option selection for Summary, Quiz, Q&A, or Study Plan
-option = st.sidebar.radio("Choose an option", ('Generate Summary', 'Generate Quiz', 'Ask a Question', 'Generate Study Plan'))
+# Sidebar option selection for Summary, Quiz, Q&A, Study Plan, Vocabulary Builder, or Topic Extraction
+option = st.sidebar.radio("Choose an option", ('Generate Summary', 'Generate Quiz', 'Ask a Question', 'Generate Study Plan', 'Vocabulary Builder', 'Topic Extraction'))
 
 # Input for asking questions (only visible when "Ask a Question" is selected)
 question_input = None
@@ -43,6 +48,28 @@ def generate_pdf(response, filename="response.pdf"):
     # Return the file path
     return filename
 
+def extract_key_terms(text):
+    """
+    Extract key terms (nouns) from the text using spaCy.
+    """
+    doc = nlp(text)
+    key_terms = [token.text for token in doc if token.pos_ == "NOUN" and not token.is_stop]
+    
+    # Count frequency of terms (for demonstration purposes, we show the top 10 most frequent terms)
+    term_frequency = Counter(key_terms)
+    most_common_terms = term_frequency.most_common(10)
+    
+    return most_common_terms
+
+def extract_topics(text):
+    """
+    Extract main topics (entities) using spaCy's Named Entity Recognition (NER).
+    """
+    doc = nlp(text)
+    topics = [ent.text for ent in doc.ents]  # Extract named entities (topics)
+    
+    return set(topics)  # Return unique entities as topics
+
 if openai_api_key:
     # Set OpenAI API key
     openai.api_key = openai_api_key
@@ -151,5 +178,27 @@ if openai_api_key:
         pdf_filename = generate_pdf(study_plan_response, filename="study_plan_response.pdf")
         st.download_button("Download Study Plan as PDF", data=open(pdf_filename, "rb").read(), file_name=pdf_filename, mime="application/pdf")
 
+    elif option == 'Vocabulary Builder':
+        # Extract key terms from the document
+        key_terms = extract_key_terms(' '.join([page['text'] for page in data]))
+        st.write("Key Terms and Definitions:")
+        for term, frequency in key_terms:
+            st.write(f"{term} - Frequency: {frequency}")
+        
+        # Generate PDF for the key terms and offer it as a download
+        pdf_filename = generate_pdf(str(key_terms), filename="key_terms.pdf")
+        st.download_button("Download Vocabulary Builder PDF", data=open(pdf_filename, "rb").read(), file_name=pdf_filename, mime="application/pdf")
+
+    elif option == 'Topic Extraction':
+        # Extract topics (named entities) from the document
+        topics = extract_topics(' '.join([page['text'] for page in data]))
+        st.write("Extracted Topics:")
+        for topic in topics:
+            st.write(topic)
+        
+        # Generate PDF for the topics and offer it as a download
+        pdf_filename = generate_pdf(str(topics), filename="topics.pdf")
+        st.download_button("Download Topics PDF", data=open(pdf_filename, "rb").read(), file_name=pdf_filename, mime="application/pdf")
+
 else:
     st.sidebar.warning("Please enter your OpenAI API Key to proceed.")