generate_presentation.py

"""
Generate PowerPoint Presentation for Profyler Project
Simple, clear slides for teacher presentation
"""

from pptx import Presentation
from pptx.util import Inches, Pt
from pptx.enum.text import PP_ALIGN, MSO_ANCHOR
from pptx.dml.color import RGBColor
import os

def create_profyler_presentation():
    """Create comprehensive PPT for Profyler project"""
    
    # Create presentation
    prs = Presentation()
    prs.slide_width = Inches(10)
    prs.slide_height = Inches(7.5)
    
    # Define color scheme (Purple gradient theme)
    PRIMARY_COLOR = RGBColor(102, 126, 234)  # #667eea
    SECONDARY_COLOR = RGBColor(118, 75, 162)  # #764ba2
    TEXT_COLOR = RGBColor(51, 51, 51)
    LIGHT_GRAY = RGBColor(107, 114, 128)
    
    def add_title_slide(title, subtitle=""):
        """Add a title slide"""
        slide = prs.slides.add_slide(prs.slide_layouts[6])  # Blank layout
        
        # Background
        background = slide.shapes.add_shape(
            1,  # Rectangle
            0, 0, prs.slide_width, prs.slide_height
        )
        background.fill.solid()
        background.fill.fore_color.rgb = RGBColor(255, 255, 255)
        background.line.fill.background()
        
        # Title
        title_box = slide.shapes.add_textbox(
            Inches(0.5), Inches(2.5), Inches(9), Inches(1.5)
        )
        title_frame = title_box.text_frame
        title_frame.text = title
        title_p = title_frame.paragraphs[0]
        title_p.font.size = Pt(54)
        title_p.font.bold = True
        title_p.font.color.rgb = PRIMARY_COLOR
        title_p.alignment = PP_ALIGN.CENTER
        
        # Subtitle
        if subtitle:
            subtitle_box = slide.shapes.add_textbox(
                Inches(0.5), Inches(4.2), Inches(9), Inches(1)
            )
            subtitle_frame = subtitle_box.text_frame
            subtitle_frame.text = subtitle
            subtitle_p = subtitle_frame.paragraphs[0]
            subtitle_p.font.size = Pt(24)
            subtitle_p.font.color.rgb = LIGHT_GRAY
            subtitle_p.alignment = PP_ALIGN.CENTER
        
        return slide
    
    def add_content_slide(title, content_items, bullet_style=True):
        """Add a content slide with bullet points or paragraphs"""
        slide = prs.slides.add_slide(prs.slide_layouts[6])
        
        # Title
        title_box = slide.shapes.add_textbox(
            Inches(0.5), Inches(0.4), Inches(9), Inches(0.8)
        )
        title_frame = title_box.text_frame
        title_frame.text = title
        title_p = title_frame.paragraphs[0]
        title_p.font.size = Pt(36)
        title_p.font.bold = True
        title_p.font.color.rgb = PRIMARY_COLOR
        
        # Content
        content_box = slide.shapes.add_textbox(
            Inches(0.8), Inches(1.5), Inches(8.4), Inches(5.5)
        )
        content_frame = content_box.text_frame
        content_frame.word_wrap = True
        
        for i, item in enumerate(content_items):
            if i > 0:
                p = content_frame.add_paragraph()
            else:
                p = content_frame.paragraphs[0]
            
            p.text = item
            p.font.size = Pt(20)
            p.font.color.rgb = TEXT_COLOR
            p.space_after = Pt(12)
            
            if bullet_style:
                p.level = 0
        
        return slide
    
    def add_two_column_slide(title, left_content, right_content):
        """Add a slide with two columns"""
        slide = prs.slides.add_slide(prs.slide_layouts[6])
        
        # Title
        title_box = slide.shapes.add_textbox(
            Inches(0.5), Inches(0.4), Inches(9), Inches(0.8)
        )
        title_frame = title_box.text_frame
        title_frame.text = title
        title_p = title_frame.paragraphs[0]
        title_p.font.size = Pt(36)
        title_p.font.bold = True
        title_p.font.color.rgb = PRIMARY_COLOR
        
        # Left column
        left_box = slide.shapes.add_textbox(
            Inches(0.5), Inches(1.5), Inches(4.5), Inches(5.5)
        )
        left_frame = left_box.text_frame
        left_frame.word_wrap = True
        
        for i, item in enumerate(left_content):
            if i > 0:
                p = left_frame.add_paragraph()
            else:
                p = left_frame.paragraphs[0]
            p.text = item
            p.font.size = Pt(18)
            p.font.color.rgb = TEXT_COLOR
            p.space_after = Pt(10)
        
        # Right column
        right_box = slide.shapes.add_textbox(
            Inches(5.2), Inches(1.5), Inches(4.3), Inches(5.5)
        )
        right_frame = right_box.text_frame
        right_frame.word_wrap = True
        
        for i, item in enumerate(right_content):
            if i > 0:
                p = right_frame.add_paragraph()
            else:
                p = right_frame.paragraphs[0]
            p.text = item
            p.font.size = Pt(18)
            p.font.color.rgb = TEXT_COLOR
            p.space_after = Pt(10)
        
        return slide
    
    # ========================================
    # SLIDE 1: TITLE SLIDE
    # ========================================
    slide1 = add_title_slide(
        "Profyler",
        "AI-Powered Research Impact Prediction System"
    )
    
    # Add authors
    authors_box = slide1.shapes.add_textbox(
        Inches(0.5), Inches(5.5), Inches(9), Inches(1)
    )
    authors_frame = authors_box.text_frame
    authors_frame.text = "Om Harsh (E23CSEU2423) & Kshitiz Yadav (E23CSEU0338)\nChitkara University"
    for p in authors_frame.paragraphs:
        p.font.size = Pt(18)
        p.font.color.rgb = TEXT_COLOR
        p.alignment = PP_ALIGN.CENTER
    
    # ========================================
    # SLIDE 2: PROJECT IDEA
    # ========================================
    idea_content = [
        "Problem: Researchers and institutions need automated tools to evaluate research impact and identify emerging trends",
        "",
        "Solution: Profyler - An AI system that:",
        "  • Aggregates publications from multiple academic databases",
        "  • Predicts citation impact using machine learning (99.91% accuracy)",
        "  • Analyzes research trends and collaboration patterns",
        "  • Provides strategic recommendations for future research",
        "",
        "Target Users: Researchers, academic institutions, funding agencies"
    ]
    add_content_slide("💡 Project Idea", idea_content, bullet_style=False)
    
    # ========================================
    # SLIDE 3: TECHNICAL APPROACH - FLOW DIAGRAM
    # ========================================
    slide3 = prs.slides.add_slide(prs.slide_layouts[6])
    
    # Title
    title_box = slide3.shapes.add_textbox(
        Inches(0.5), Inches(0.4), Inches(9), Inches(0.8)
    )
    title_frame = title_box.text_frame
    title_frame.text = "⚙️ Technical Approach - System Flow"
    title_p = title_frame.paragraphs[0]
    title_p.font.size = Pt(36)
    title_p.font.bold = True
    title_p.font.color.rgb = PRIMARY_COLOR
    
    # Flow diagram boxes
    flow_steps = [
        ("1. Data Collection", "Fetch from Semantic Scholar,\nDBLP, CrossRef APIs"),
        ("2. Data Merging", "Intelligent deduplication\nusing DOI & title matching"),
        ("3. Feature Engineering", "Extract 36 features:\nTemporal, Venue, Collaboration,\nContent, Domain, Innovation"),
        ("4. ML Pipeline", "SMOTE balancing →\nRandom Forest + Gradient Boost\n→ Ensemble Prediction"),
        ("5. AI Analysis", "Trend analysis, Impact prediction,\nStrategic recommendations"),
        ("6. Presentation", "Web interface with charts,\nExcel/Word reports")
    ]
    
    y_pos = 1.5
    for step, description in flow_steps:
        # Box
        box = slide3.shapes.add_shape(
            1,  # Rectangle
            Inches(1.5), Inches(y_pos), Inches(7), Inches(0.8)
        )
        box.fill.solid()
        box.fill.fore_color.rgb = RGBColor(240, 240, 255)
        box.line.color.rgb = PRIMARY_COLOR
        box.line.width = Pt(2)
        
        # Text
        text_frame = box.text_frame
        text_frame.text = f"{step}\n{description}"
        text_frame.word_wrap = True
        
        p = text_frame.paragraphs[0]
        p.font.size = Pt(14)
        p.font.bold = True
        p.font.color.rgb = PRIMARY_COLOR
        
        if len(text_frame.paragraphs) > 1:
            for para in text_frame.paragraphs[1:]:
                para.font.size = Pt(11)
                para.font.color.rgb = TEXT_COLOR
        
        # Arrow (except for last step)
        if y_pos < 6.5:
            arrow = slide3.shapes.add_shape(
                12,  # Down arrow
                Inches(4.8), Inches(y_pos + 0.85), Inches(0.4), Inches(0.25)
            )
            arrow.fill.solid()
            arrow.fill.fore_color.rgb = PRIMARY_COLOR
            arrow.line.fill.background()
        
        y_pos += 1.0
    
    # ========================================
    # SLIDE 4: TECH STACK
    # ========================================
    tech_left = [
        "Backend:",
        "• Flask 3.0.3 (REST API)",
        "• Python 3.10",
        "",
        "Frontend:",
        "• Streamlit 1.29.0",
        "• Plotly (Charts)",
        "",
        "Data Sources:",
        "• Semantic Scholar API",
        "• DBLP API",
        "• CrossRef API"
    ]
    
    tech_right = [
        "Machine Learning:",
        "• scikit-learn 1.3.2",
        "• imbalanced-learn (SMOTE)",
        "• pandas, numpy",
        "",
        "Libraries:",
        "• requests (API calls)",
        "• openpyxl (Excel export)",
        "• python-docx (Word export)",
        "• aiohttp (async fetching)",
        "",
        "Deployment:",
        "• Local servers (development)",
        "• Production-ready architecture"
    ]
    
    add_two_column_slide("🛠️ Technology Stack", tech_left, tech_right)
    
    # ========================================
    # SLIDE 5: NOVELTY
    # ========================================
    novelty_content = [
        "1. Multi-Source Aggregation",
        "   First system to intelligently merge 3 academic databases with 460+ publications per researcher",
        "",
        "2. Class Imbalance Handling",
        "   SMOTE optimization achieving 99.91% accuracy on severely imbalanced data (84.6% zero citations)",
        "",
        "3. Comprehensive Feature Engineering",
        "   36 domain-specific features spanning 6 categories (vs. traditional h-index only)",
        "",
        "4. Ensemble Learning Approach",
        "   Random Forest + Gradient Boosting with soft voting (99.93% accuracy each)",
        "",
        "5. Real-time Research Profiling",
        "   Interactive web interface with AI-powered insights and trend predictions"
    ]
    add_content_slide("🌟 Novelty & Innovation", novelty_content, bullet_style=False)
    
    # ========================================
    # SLIDE 6: FEASIBILITY & VIABILITY
    # ========================================
    feasibility_left = [
        "Technical Feasibility:",
        "",
        "✅ Fully Implemented",
        "• Working system with Flask + Streamlit",
        "• Trained models (99.91% accuracy)",
        "• Multi-source data aggregation",
        "",
        "✅ Scalable Architecture",
        "• Async API calls",
        "• Configurable limits (10k pubs)",
        "• Memory-efficient SMOTE (30%)",
        "",
        "✅ Production Ready",
        "• Error handling & SSL fixes",
        "• Auto-save functionality",
        "• Export to CSV/Excel/Word"
    ]
    
    feasibility_right = [
        "Economic Viability:",
        "",
        "💰 Low Cost",
        "• Free academic APIs",
        "• Open-source libraries",
        "• Minimal infrastructure",
        "",
        "📈 High Value",
        "• Saves researcher time",
        "• Institutional decision support",
        "• Funding allocation insights",
        "",
        "🎯 Market Potential",
        "• Universities & research labs",
        "• Funding agencies",
        "• Academic publishers",
        "• Citation analysis companies"
    ]
    
    add_two_column_slide("✅ Feasibility & Viability", feasibility_left, feasibility_right)
    
    # ========================================
    # SLIDE 7: IMPACT & BENEFITS
    # ========================================
    impact_content = [
        "Academic Impact:",
        "• 78.43% improvement over baseline (21.48% → 99.91% accuracy)",
        "• Publishable in top-tier conferences (ACM KDD, AAAI, WSDM)",
        "",
        "Practical Benefits:",
        "• Automated Research Profiling: Generate comprehensive profiles in minutes",
        "• Impact Prediction: Identify high-potential papers before they gain citations",
        "• Trend Analysis: Discover emerging research topics and collaboration opportunities",
        "• Strategic Guidance: AI-powered recommendations for future research directions",
        "",
        "Stakeholder Benefits:",
        "• Researchers: Understand impact, find collaborators, identify gaps",
        "• Institutions: Evaluate faculty, allocate resources, track research growth",
        "• Funding Agencies: Assess grant applications, predict research outcomes"
    ]
    add_content_slide("🎯 Impact & Benefits", impact_content, bullet_style=False)
    
    # ========================================
    # SLIDE 8: RESULTS & METRICS
    # ========================================
    results_left = [
        "Performance Metrics:",
        "",
        "🎯 Accuracy: 99.91%",
        "🎯 Precision: 99.91%",
        "🎯 Recall: 99.91%",
        "🎯 F1-Score: 99.90%",
        "",
        "Dataset:",
        "• 10,000 publications (training)",
        "• 70-15-15 train-val-test split",
        "• Multiple domains & years",
        "• Real citations (0 to 1000+)"
    ]
    
    results_right = [
        "Model Comparison:",
        "",
        "Baseline (Heuristic):",
        "  21.48% accuracy",
        "",
        "Random Forest:",
        "  99.93% accuracy",
        "",
        "Gradient Boosting:",
        "  99.93% accuracy",
        "",
        "Ensemble (Our Model):",
        "  99.93% accuracy",
        "",
        "Test Case:",
        "  Andrew Ng: 460 pubs aggregated"
    ]
    
    add_two_column_slide("📊 Results & Metrics", results_left, results_right)
    
    # ========================================
    # SLIDE 9: DEMO SCREENSHOTS
    # ========================================
    demo_content = [
        "System Features:",
        "",
        "🔍 Input Interface",
        "   • Single/multi-author search",
        "   • CSV/Excel/BibTeX upload",
        "   • Advanced filters (year, type, affiliation)",
        "",
        "📊 Analysis Dashboard",
        "   • Key metrics: H-index, i10-index, total publications",
        "   • Interactive charts: Timeline, impact distribution, trends",
        "   • Publication table with filters",
        "",
        "🤖 AI Insights",
        "   • Impact predictions per paper",
        "   • Research trend analysis (upward/stable/downward)",
        "   • Collaboration recommendations",
        "   • Strategic insights for future research"
    ]
    add_content_slide("🖥️ System Demo", demo_content, bullet_style=False)
    
    # ========================================
    # SLIDE 10: FUTURE ENHANCEMENTS
    # ========================================
    future_content = [
        "Planned Improvements:",
        "",
        "🔬 Advanced ML",
        "   • Deep learning with BERT embeddings for content analysis",
        "   • Network-based features from co-authorship graphs",
        "   • Real-time citation tracking with incremental updates",
        "",
        "📈 Extended Analysis",
        "   • Cross-domain validation (CS, Medicine, Physics)",
        "   • Temporal robustness (train on old data, test on new)",
        "   • Author disambiguation and ORCID integration",
        "",
        "🌐 Deployment",
        "   • Cloud deployment (AWS/Azure/Heroku)",
        "   • API for institutional integration",
        "   • Mobile application for researchers"
    ]
    add_content_slide("🚀 Future Enhancements", future_content, bullet_style=False)
    
    # ========================================
    # SLIDE 11: THANK YOU
    # ========================================
    slide_thanks = add_title_slide(
        "Thank You!",
        "Questions & Discussion"
    )
    
    # Add contact info
    contact_box = slide_thanks.shapes.add_textbox(
        Inches(0.5), Inches(5.5), Inches(9), Inches(1.5)
    )
    contact_frame = contact_box.text_frame
    contact_frame.text = "Om Harsh - E23CSEU2423\nKshitiz Yadav - E23CSEU0338\n\nSmart India Hackathon 2024 | Chitkara University"
    for p in contact_frame.paragraphs:
        p.font.size = Pt(16)
        p.font.color.rgb = TEXT_COLOR
        p.alignment = PP_ALIGN.CENTER
    
    # Save presentation
    output_path = os.path.join(os.path.dirname(__file__), "Profyler_Presentation.pptx")
    prs.save(output_path)
    print(f"✅ Presentation created successfully: {output_path}")
    return output_path

if __name__ == "__main__":
    try:
        path = create_profyler_presentation()
        print(f"\n📊 PowerPoint presentation ready!")
        print(f"📁 Location: {path}")
        print(f"\n🎯 Total slides: 11")
        print(f"   1. Title Slide")
        print(f"   2. Project Idea")
        print(f"   3. Technical Approach (Flow Diagram)")
        print(f"   4. Technology Stack")
        print(f"   5. Novelty & Innovation")
        print(f"   6. Feasibility & Viability")
        print(f"   7. Impact & Benefits")
        print(f"   8. Results & Metrics")
        print(f"   9. System Demo")
        print(f"   10. Future Enhancements")
        print(f"   11. Thank You")
    except Exception as e:
        print(f"❌ Error creating presentation: {e}")
        import traceback
        traceback.print_exc()