Create app.py

app.py

@@ -0,0 +1,117 @@
+import torch
+import gradio as gr
+import numpy as np
+from transformers import AutoTokenizer, AutoModelForSequenceClassification
+
+class ThreatAnalyzer:
+    def __init__(self):
+        # Load pretrained model for threat detection
+        model_name = 'distilbert-base-uncased-finetuned-sst-2-english'
+        self.tokenizer = AutoTokenizer.from_pretrained(model_name)
+        self.model = AutoModelForSequenceClassification.from_pretrained(model_name)
+
+    def analyze_threat(self, input_text):
+        """
+        Comprehensive threat analysis with recommendations
+        """
+        # Detect threat probability
+        inputs = self.tokenizer(input_text, return_tensors="pt", truncation=True, padding=True)
+        outputs = self.model(**inputs)
+        predictions = torch.nn.functional.softmax(outputs.logits, dim=-1)
+        threat_score = predictions[0][1].item()
+
+        # Categorize threat types
+        threat_patterns = {
+            "phishing": ["verify", "urgent", "suspend", "account", "click here"],
+            "social_engineering": ["tech support", "password", "credentials", "urgent"],
+            "malware": ["download", "patch", "update", "critical"],
+            "impersonation": ["admin", "support", "verify identity"]
+        }
+
+        # Analyze input for specific threat patterns
+        detected_patterns = []
+        for threat_type, keywords in threat_patterns.items():
+            if any(keyword.lower() in input_text.lower() for keyword in keywords):
+                detected_patterns.append(threat_type)
+
+        # Generate recommendations based on threat analysis
+        recommendations = []
+        if threat_score > 0.7:
+            recommendations.extend([
+                "🚨 High Risk: Do NOT interact with this message",
+                "Immediately report to IT security",
+                "Do not click any links or download attachments"
+            ])
+        elif threat_score > 0.4:
+            recommendations.extend([
+                "⚠️ Potential Threat Detected",
+                "Verify the source independently",
+                "Contact sender through official channels"
+            ])
+
+        # If specific threat patterns detected, add specific advice
+        if "phishing" in detected_patterns:
+            recommendations.append("Phishing Indicator: Check sender's email address carefully")
+        if "social_engineering" in detected_patterns:
+            recommendations.append("Social Engineering Alert: Never share personal credentials")
+        if "malware" in detected_patterns:
+            recommendations.append("Malware Risk: Scan with updated antivirus before opening")
+
+        return {
+            "threat_score": threat_score,
+            "threat_level": self._classify_threat_level(threat_score),
+            "detected_patterns": detected_patterns,
+            "recommendations": recommendations
+        }
+
+    def _classify_threat_level(self, score):
+        """Classify threat level"""
+        if score > 0.8: return "CRITICAL THREAT"
+        if score > 0.6: return "HIGH THREAT"
+        if score > 0.4: return "MODERATE THREAT"
+        if score > 0.2: return "LOW THREAT"
+        return "NO SIGNIFICANT THREAT"
+
+def launch_threat_analysis_demo():
+    """Create Gradio interface for threat analysis"""
+    analyzer = ThreatAnalyzer()
+
+    def analyze_input(input_text):
+        try:
+            result = analyzer.analyze_threat(input_text)
+
+            # Format output for clear presentation
+            output = f"""
+            🔍 Threat Analysis Results:
+
+            Threat Score: {result['threat_score']:.2%}
+            Threat Level: {result['threat_level']}
+
+            Detected Threat Patterns:
+            {', '.join(result['detected_patterns']) or 'No specific patterns identified'}
+
+            Recommendations:
+            {chr(10).join('• ' + rec for rec in result['recommendations'])}
+
+            💡 Always verify suspicious communications
+            """
+            return output
+        except Exception as e:
+            return f"Error in analysis: {str(e)}"
+
+    # Gradio interface
+    with gr.Blocks() as demo:
+        gr.Markdown("# Cybersecurity Threat Analysis")
+
+        input_text = gr.Textbox(label="Enter suspicious text or message")
+        output_text = gr.Textbox(label="Threat Analysis Results")
+
+        analyze_btn = gr.Button("Analyze Threat")
+        analyze_btn.click(fn=analyze_input, inputs=input_text, outputs=output_text)
+
+    return demo
+
+# Run the demo
+if __name__ == "__main__":
+    demo = launch_threat_analysis_demo()
+    demo.launch(debug=True, share=True)