Integrate teenager protection tools to social media analyzer.

This commit introduces a new set of tools focused on protecting teenagers from online threats. The new features include detection for:
- Cyberbullying
- Inappropriate content
- Privacy risks (oversharing)

A new module, `teen_protection.py`, has been added to house the analysis logic. The main application has been updated to include a new menu for these tools. Heuristics have been expanded with relevant keywords and weights.

Unit tests for the new functionality have been added and integrated into the existing test suite, and all tests are passing.

Author: google-labs-jules[bot]

social_media_analyzer/heuristics.py

@@ -122,6 +122,37 @@
     "bank transfer", "wire details", "account details", "iban", "swift code", "bic"
 ]
 
+# --- Teenager Protection Heuristics ---
+
+# Keywords/phrases related to cyberbullying
+CYBERBULLYING_KEYWORDS = [
+    "loser", "stupid", "idiot", "hate you", "ugly", "fat",
+    "kill yourself", "kys", "go die", "nobody likes you", "freak",
+    "weirdo", "everyone hates you", "you're worthless", "pathetic",
+    "troll", "noob", "poser", "wannabe", "go away",
+    "social reject", "outcast", "misfit", "dork", "nerd"
+]
+
+# Keywords/phrases related to inappropriate content (sexual, violent, etc.)
+INAPPROPRIATE_CONTENT_KEYWORDS = [
+    # Sexually suggestive
+    "nude", "sexting", "send nudes", "horny", "slut", "whore", "dick", "pussy",
+    "porn", "sexy pic", "private parts", "hook up",
+    # Violence
+    "kill", "murder", "blood", "gun", "knife", "fight me",
+    "i will hurt you", "beat you up", "gonna get you",
+    # Drugs/Alcohol
+    "drugs", "weed", "cocaine", "pills", "get high", "drunk", "wasted"
+]
+
+# Keywords/phrases indicating oversharing of personal information
+PRIVACY_RISK_KEYWORDS = [
+    "my address is", "i live at", "my phone number is", "call me at",
+    "my full name is", "my school is", "i go to [school_name]",
+    "my mom's name is", "my dad's name is",
+    "i'm home alone", "parents are out", "my password is"
+]
+
 
 # --- Fake News Heuristics ---
 
@@ -240,6 +271,10 @@ def generate_suspicious_url_patterns(legitimate_domains):
     "PHONE_NUMBER_UNSOLICITED": 1.0,
     "SUSPICIOUS_URL_PATTERN": 3.0, # High weight for matching a suspicious URL pattern
     "GOOGLE_SAFE_BROWSING_HIT": 10.0, # Very high weight for a positive Google Safe Browsing match
+    # Teenager Protection Weights
+    "CYBERBULLYING": 2.5,
+    "INAPPROPRIATE_CONTENT": 3.0,
+    "PRIVACY_RISK": 3.5,
 }
 
 if __name__ == '__main__':

social_media_analyzer/main.py

@@ -2,6 +2,7 @@
 from . import fake_profile_detector
 from . import scam_detector
 from . import fake_news_detector
+from . import teen_protection
 
 def get_api_key():
     """Gets the Google API key from environment variables."""
@@ -130,6 +131,48 @@ def analyze_social_media(api_key):
         except ValueError:
             print("Invalid input. Please enter a number.")
 
+def analyze_for_teen_risks():
+    """Handles analysis for risks relevant to teenagers."""
+    print("\n--- Teenager Protection Tools ---")
+    print("Select the type of analysis you want to perform:")
+    print("1. Analyze text for Cyberbullying")
+    print("2. Analyze text for Inappropriate Content")
+    print("3. Analyze text for Privacy Risks (Oversharing)")
+
+    try:
+        choice = int(input("Enter your choice (1-3): "))
+        if choice not in [1, 2, 3]:
+            print("Invalid choice. Please try again.")
+            return
+    except ValueError:
+        print("Invalid input. Please enter a number.")
+        return
+
+    text_to_analyze = input("Please paste the text you want to analyze: ").strip()
+    if not text_to_analyze:
+        print("No text entered.")
+        return
+
+    result = {}
+    if choice == 1:
+        print("\n--- Analyzing for Cyberbullying ---")
+        result = teen_protection.analyze_for_cyberbullying(text_to_analyze)
+    elif choice == 2:
+        print("\n--- Analyzing for Inappropriate Content ---")
+        result = teen_protection.analyze_for_inappropriate_content(text_to_analyze)
+    elif choice == 3:
+        print("\n--- Analyzing for Privacy Risks ---")
+        result = teen_protection.analyze_for_privacy_risks(text_to_analyze)
+
+    print(f"Score: {result['score']} (Higher is more suspicious)")
+    if result['indicators_found']:
+        print("Indicators Found:")
+        for indicator in result['indicators_found']:
+            print(f"- {indicator}")
+    else:
+        print("No specific risk indicators were found.")
+
+
 def main():
     """Main function to run the security analyzer."""
     api_key = get_api_key()
@@ -145,17 +188,20 @@ def main():
         print("1. Analyze a Social Media Platform")
         print("2. Analyze a Website URL for Scams")
         print("3. Analyze a News URL for Fake News")
-        print("4. Exit")
+        print("4. Teenager Protection Tools")
+        print("5. Exit")
 
         try:
-            choice = int(input("Enter your choice (1-4): "))
+            choice = int(input("Enter your choice (1-5): "))
             if choice == 1:
                 analyze_social_media(api_key)
             elif choice == 2:
                 analyze_website_url(api_key)
             elif choice == 3:
                 analyze_news_url()
             elif choice == 4:
+                analyze_for_teen_risks()
+            elif choice == 5:
                 print("Exiting. Stay safe!")
                 break
             else:

social_media_analyzer/teen_protection.py

@@ -0,0 +1,58 @@
+from .heuristics import (
+    CYBERBULLYING_KEYWORDS,
+    INAPPROPRIATE_CONTENT_KEYWORDS,
+    PRIVACY_RISK_KEYWORDS,
+    HEURISTIC_WEIGHTS
+)
+
+def analyze_text_for_teen_risks(text, analysis_type):
+    """
+    Analyzes text for a specific type of risk to teenagers.
+
+    :param text: The text content to analyze.
+    :param analysis_type: The type of analysis to perform ('cyberbullying',
+                          'inappropriate_content', 'privacy_risk').
+    :return: A dictionary with the score and indicators found.
+    """
+    if not text:
+        return {"score": 0.0, "indicators_found": []}
+
+    text_lower = text.lower()
+    score = 0.0
+    indicators_found = []
+
+    keyword_map = {
+        'cyberbullying': ('CYBERBULLYING', CYBERBULLYING_KEYWORDS),
+        'inappropriate_content': ('INAPPROPRIATE_CONTENT', INAPPROPRIATE_CONTENT_KEYWORDS),
+        'privacy_risk': ('PRIVACY_RISK', PRIVACY_RISK_KEYWORDS),
+    }
+
+    if analysis_type not in keyword_map:
+        return {"error": "Invalid analysis type specified."}
+
+    category, keywords = keyword_map[analysis_type]
+    weight = HEURISTIC_WEIGHTS.get(category.upper(), 1.0)
+
+    for keyword in keywords:
+        if keyword in text_lower:
+            message = f"Detected potential {category.replace('_', ' ').lower()} keyword: '{keyword}'"
+            if message not in indicators_found:
+                indicators_found.append(message)
+                score += weight
+
+    return {
+        "score": round(score, 2),
+        "indicators_found": indicators_found
+    }
+
+def analyze_for_cyberbullying(text):
+    """Analyzes text for signs of cyberbullying."""
+    return analyze_text_for_teen_risks(text, 'cyberbullying')
+
+def analyze_for_inappropriate_content(text):
+    """Analyzes text for inappropriate content."""
+    return analyze_text_for_teen_risks(text, 'inappropriate_content')
+
+def analyze_for_privacy_risks(text):
+    """Analyzes text for privacy risks (oversharing)."""
+    return analyze_text_for_teen_risks(text, 'privacy_risk')

social_media_analyzer/test_runner.py

@@ -1,6 +1,7 @@
 import unittest
 from unittest.mock import patch, Mock
-from social_media_analyzer.scam_detector import analyze_text_for_scams
+from .scam_detector import analyze_text_for_scams
+from .test_teen_protection import TestTeenProtection
 
 def run_manual_tests():
     # Example Usage
@@ -91,11 +92,13 @@ def test_google_safe_browsing_clean(self, mock_post):
 if __name__ == '__main__':
     run_manual_tests()
     # Run unit tests
-    suite = unittest.TestSuite()
-    suite.addTest(unittest.makeSuite(TestScamDetector))
+    scam_suite = unittest.makeSuite(TestScamDetector)
+    teen_suite = unittest.makeSuite(TestTeenProtection)
+    all_tests = unittest.TestSuite([scam_suite, teen_suite])
+
     runner = unittest.TextTestRunner()
-    print("\n--- Running Unit Tests for Google Safe Browsing Integration ---")
-    result = runner.run(suite)
+    print("\n--- Running All Unit Tests ---")
+    result = runner.run(all_tests)
     if result.wasSuccessful():
         print("All tests passed!")
     else:

social_media_analyzer/test_teen_protection.py

@@ -0,0 +1,70 @@
+import unittest
+from .teen_protection import (
+    analyze_for_cyberbullying,
+    analyze_for_inappropriate_content,
+    analyze_for_privacy_risks
+)
+
+class TestTeenProtection(unittest.TestCase):
+
+    def test_cyberbullying(self):
+        """Test the cyberbullying detection."""
+        # Test case with bullying keywords
+        text1 = "You are such a loser and an idiot."
+        result1 = analyze_for_cyberbullying(text1)
+        self.assertGreater(result1['score'], 0)
+        self.assertIn("Detected potential cyberbullying keyword: 'loser'", result1['indicators_found'])
+        self.assertIn("Detected potential cyberbullying keyword: 'idiot'", result1['indicators_found'])
+
+        # Test case with no bullying keywords
+        text2 = "Have a great day!"
+        result2 = analyze_for_cyberbullying(text2)
+        self.assertEqual(result2['score'], 0)
+        self.assertEqual(len(result2['indicators_found']), 0)
+
+    def test_inappropriate_content(self):
+        """Test the inappropriate content detection."""
+        # Test case with inappropriate keywords
+        text1 = "Don't send nudes or talk about drugs."
+        result1 = analyze_for_inappropriate_content(text1)
+        self.assertGreater(result1['score'], 0)
+        self.assertIn("Detected potential inappropriate content keyword: 'send nudes'", result1['indicators_found'])
+        self.assertIn("Detected potential inappropriate content keyword: 'drugs'", result1['indicators_found'])
+
+        # Test case with no inappropriate keywords
+        text2 = "This is a perfectly normal conversation."
+        result2 = analyze_for_inappropriate_content(text2)
+        self.assertEqual(result2['score'], 0)
+        self.assertEqual(len(result2['indicators_found']), 0)
+
+    def test_privacy_risks(self):
+        """Test the privacy risk detection."""
+        # Test case with privacy risk keywords
+        text1 = "My address is 123 Main St and my phone number is 555-1234."
+        result1 = analyze_for_privacy_risks(text1)
+        self.assertGreater(result1['score'], 0)
+        self.assertIn("Detected potential privacy risk keyword: 'my address is'", result1['indicators_found'])
+        self.assertIn("Detected potential privacy risk keyword: 'my phone number is'", result1['indicators_found'])
+
+        # Test case with no privacy risk keywords
+        text2 = "I like to talk about my hobbies."
+        result2 = analyze_for_privacy_risks(text2)
+        self.assertEqual(result2['score'], 0)
+        self.assertEqual(len(result2['indicators_found']), 0)
+
+    def test_empty_input(self):
+        """Test empty input for all analysis types."""
+        result_cb = analyze_for_cyberbullying("")
+        self.assertEqual(result_cb['score'], 0)
+        self.assertEqual(len(result_cb['indicators_found']), 0)
+
+        result_ic = analyze_for_inappropriate_content("")
+        self.assertEqual(result_ic['score'], 0)
+        self.assertEqual(len(result_ic['indicators_found']), 0)
+
+        result_pr = analyze_for_privacy_risks("")
+        self.assertEqual(result_pr['score'], 0)
+        self.assertEqual(len(result_pr['indicators_found']), 0)
+
+if __name__ == '__main__':
+    unittest.main()