Update app.py

app.py

@@ -2,9 +2,10 @@ import streamlit as st
 import nltk
 import spacy
 import matplotlib.pyplot as plt
-from transformers import pipeline
+from transformers import pipeline, AutoTokenizer, AutoModelForSequenceClassification
 import random
 import subprocess
+import torch
 
 # Ensure spaCy model is installed
 try:
@@ -13,11 +14,17 @@ except OSError:
     subprocess.run(["python", "-m", "spacy", "download", "en_core_web_sm"])
     nlp = spacy.load("en_core_web_sm")
 
-# Load NLP models
+# Ensure NLTK resources are available
 nltk.download("vader_lexicon")
 from nltk.sentiment import SentimentIntensityAnalyzer
 sia = SentimentIntensityAnalyzer()
 
+# Load multilingual emotion detection model
+model_name = "joeddav/xlm-roberta-large-xnli"
+tokenizer = AutoTokenizer.from_pretrained(model_name)
+emotion_model = AutoModelForSequenceClassification.from_pretrained(model_name)
+
+# Load English emotion detection pipeline
 emotion_pipeline = pipeline("text-classification", model="j-hartmann/emotion-english-distilroberta-base", return_all_scores=True)
 
 # Sample texts
@@ -27,10 +34,21 @@ sample_texts = [
     "AI-driven education tools can personalize the learning experience for students."
 ]
 
+# Supported languages for multilingual sentiment analysis
+supported_languages = {
+    "English": "en",
+    "Dutch": "nl",
+    "French": "fr",
+    "German": "de"
+}
+
 # Streamlit UI
 st.title("📖 AI-Powered Adaptive Reading Engagement")
 st.write("Analyze how users engage with digital reading using AI-powered insights.")
 
+# Language selection
+selected_language = st.selectbox("Select a language:", list(supported_languages.keys()))
+
 # Use session state to store text input
 if "user_text" not in st.session_state:
     st.session_state.user_text = ""
@@ -59,11 +77,40 @@ if text_option == "Choose Your Own Text":
 if text_option == "Choose Your Own Text" and word_count <= MAX_WORDS:
     st.session_state.user_text = text
 
+# Function to generate AI-driven feedback
+def generate_feedback(sentiment_score, top_emotion):
+    if sentiment_score['pos'] > 0.6:
+        return "😊 Your reading is positive! This text might boost engagement and motivation."
+    elif sentiment_score['neg'] > 0.6:
+        return "😔 The text seems to have a strong negative tone. Consider balancing it with positive information."
+    elif top_emotion in ["anger", "disgust"]:
+        return "😡 This text might evoke strong emotions. Reflect on how it makes you feel and why."
+    elif top_emotion in ["joy", "optimism"]:
+        return "🎉 This is an engaging and uplifting text. Keep exploring such content!"
+    else:
+        return "🤔 The text is neutral. You might want to explore different perspectives."
+
 # Sentiment Analysis
 if st.button("Analyze Engagement"):
     if text and (text_option != "Choose Your Own Text" or word_count <= MAX_WORDS):
+        # Sentiment Analysis
         sentiment_score = sia.polarity_scores(text)
-        emotion_results = emotion_pipeline(text)
+
+        # English Emotion Detection
+        if selected_language == "English":
+            emotion_results = emotion_pipeline(text)
+            top_emotion = max(emotion_results[0], key=lambda x: x['score'])['label']
+            emotion_scores = {e['label']: e['score'] for e in emotion_results[0]}
+        else:
+            # Multilingual Sentiment Analysis
+            inputs = tokenizer(text, return_tensors="pt", truncation=True)
+            with torch.no_grad():
+                outputs = emotion_model(**inputs)
+            probabilities = torch.nn.functional.softmax(outputs.logits, dim=-1)
+            top_emotion_idx = torch.argmax(probabilities).item()
+            emotion_labels = ["neutral", "entailment", "contradiction"]
+            top_emotion = emotion_labels[top_emotion_idx]
+            emotion_scores = {top_emotion: probabilities[0][top_emotion_idx].item()}
 
         # Display Sentiment
         st.subheader("📊 Sentiment Analysis")
@@ -71,14 +118,21 @@ if st.button("Analyze Engagement"):
 
         # Display Emotion
         st.subheader("🎭 Emotion Detection")
-        top_emotion = max(emotion_results[0], key=lambda x: x['score'])
-        st.write(f"Detected Emotion: **{top_emotion['label']}** (Confidence: {top_emotion['score']:.2f})")
+        st.write(f"Top Emotion: **{top_emotion.capitalize()}**")
+
+        # Display Multiple Emotion Scores
+        st.subheader("🔍 Expanded Emotion Insights")
+        for emotion, score in emotion_scores.items():
+            st.write(f"**{emotion.capitalize()}**: {score:.2f}")
+
+        # AI-Generated Feedback
+        st.subheader("💡 AI-Generated Feedback")
+        feedback = generate_feedback(sentiment_score, top_emotion)
+        st.write(feedback)
 
         # Visualization
-        labels = [e['label'] for e in emotion_results[0]]
-        scores = [e['score'] for e in emotion_results[0]]
         fig, ax = plt.subplots()
-        ax.bar(labels, scores)
+        ax.bar(emotion_scores.keys(), emotion_scores.values())
         st.pyplot(fig)
     elif text_option == "Choose Your Own Text" and word_count > MAX_WORDS:
         st.warning("⚠️ Please reduce the text length to analyze.")