Spaces:
Running
Running
| <html lang="en"> | |
| <head> | |
| <meta charset="UTF-8"> | |
| <meta name="viewport" content="width=device-width, initial-scale=1.0"> | |
| <title>From Data to Defense — NIDS Survey Results</title> | |
| <link href="https://fonts.googleapis.com/css2?family=Share+Tech+Mono&family=Sora:wght@300;400;600;700;800&display=swap" rel="stylesheet"> | |
| <style> | |
| :root { | |
| --bg: #040c14; | |
| --bg2: #071622; | |
| --bg3: #0a1f2e; | |
| --cyan: #00d4ff; | |
| --cyan2: #00a8cc; | |
| --green: #00ff9d; | |
| --orange: #ff6b35; | |
| --yellow: #ffd700; | |
| --red: #ff3d5a; | |
| --text: #d4eaf7; | |
| --muted: #5a7a8a; | |
| --border: rgba(0,212,255,0.15); | |
| --mono: 'Share Tech Mono', monospace; | |
| --sans: 'Sora', sans-serif; | |
| } | |
| * { margin: 0; padding: 0; box-sizing: border-box; } | |
| html { scroll-behavior: smooth; } | |
| body { | |
| background: var(--bg); | |
| color: var(--text); | |
| font-family: var(--sans); | |
| overflow-x: hidden; | |
| cursor: default; | |
| } | |
| /* ── SCANLINE OVERLAY ── */ | |
| body::before { | |
| content: ''; | |
| position: fixed; | |
| inset: 0; | |
| background: repeating-linear-gradient(0deg, transparent, transparent 2px, rgba(0,0,0,0.07) 2px, rgba(0,0,0,0.07) 4px); | |
| pointer-events: none; | |
| z-index: 9999; | |
| } | |
| /* ── GRID BACKGROUND ── */ | |
| .grid-bg { | |
| position: fixed; | |
| inset: 0; | |
| background-image: | |
| linear-gradient(rgba(0,212,255,0.04) 1px, transparent 1px), | |
| linear-gradient(90deg, rgba(0,212,255,0.04) 1px, transparent 1px); | |
| background-size: 60px 60px; | |
| pointer-events: none; | |
| z-index: 0; | |
| } | |
| /* ── NAV ── */ | |
| nav { | |
| position: fixed; | |
| top: 0; left: 0; right: 0; | |
| z-index: 100; | |
| background: rgba(4,12,20,0.92); | |
| backdrop-filter: blur(12px); | |
| border-bottom: 1px solid var(--border); | |
| padding: 0 2rem; | |
| display: flex; | |
| align-items: center; | |
| justify-content: space-between; | |
| height: 56px; | |
| } | |
| .nav-logo { | |
| font-family: var(--mono); | |
| font-size: 0.8rem; | |
| color: var(--cyan); | |
| letter-spacing: 2px; | |
| text-transform: uppercase; | |
| } | |
| .nav-links { | |
| display: flex; | |
| gap: 2rem; | |
| list-style: none; | |
| } | |
| .nav-links a { | |
| font-family: var(--mono); | |
| font-size: 0.72rem; | |
| color: var(--muted); | |
| text-decoration: none; | |
| letter-spacing: 1px; | |
| text-transform: uppercase; | |
| transition: color 0.2s; | |
| } | |
| .nav-links a:hover { color: var(--cyan); } | |
| /* ── SECTIONS ── */ | |
| section { | |
| position: relative; | |
| z-index: 1; | |
| padding: 6rem 2rem; | |
| max-width: 1200px; | |
| margin: 0 auto; | |
| } | |
| /* ── HERO ── */ | |
| #hero { | |
| min-height: 100vh; | |
| display: flex; | |
| flex-direction: column; | |
| justify-content: center; | |
| max-width: 100%; | |
| padding: 6rem 4rem; | |
| position: relative; | |
| overflow: hidden; | |
| } | |
| .hero-tag { | |
| font-family: var(--mono); | |
| font-size: 0.72rem; | |
| color: var(--cyan); | |
| letter-spacing: 4px; | |
| text-transform: uppercase; | |
| margin-bottom: 1.5rem; | |
| opacity: 0; | |
| animation: fadeUp 0.8s 0.2s forwards; | |
| } | |
| h1 { | |
| font-size: clamp(2.5rem, 6vw, 5.5rem); | |
| font-weight: 800; | |
| line-height: 1.05; | |
| letter-spacing: -2px; | |
| max-width: 900px; | |
| opacity: 0; | |
| animation: fadeUp 0.8s 0.4s forwards; | |
| } | |
| h1 span { color: var(--cyan); } | |
| .hero-sub { | |
| margin-top: 2rem; | |
| font-size: 1rem; | |
| color: var(--muted); | |
| max-width: 600px; | |
| line-height: 1.7; | |
| opacity: 0; | |
| animation: fadeUp 0.8s 0.6s forwards; | |
| } | |
| .hero-stats { | |
| display: flex; | |
| gap: 3rem; | |
| margin-top: 4rem; | |
| opacity: 0; | |
| animation: fadeUp 0.8s 0.8s forwards; | |
| } | |
| .stat-item { text-align: left; } | |
| .stat-num { | |
| font-family: var(--mono); | |
| font-size: 2.8rem; | |
| color: var(--cyan); | |
| line-height: 1; | |
| } | |
| .stat-label { | |
| font-size: 0.72rem; | |
| color: var(--muted); | |
| letter-spacing: 2px; | |
| text-transform: uppercase; | |
| margin-top: 0.3rem; | |
| } | |
| /* Glowing orb */ | |
| .hero-orb { | |
| position: absolute; | |
| right: -100px; | |
| top: 50%; | |
| transform: translateY(-50%); | |
| width: 600px; | |
| height: 600px; | |
| border-radius: 50%; | |
| background: radial-gradient(circle, rgba(0,212,255,0.08) 0%, transparent 70%); | |
| animation: pulse 4s ease-in-out infinite; | |
| } | |
| @keyframes pulse { | |
| 0%, 100% { transform: translateY(-50%) scale(1); opacity: 0.6; } | |
| 50% { transform: translateY(-50%) scale(1.1); opacity: 1; } | |
| } | |
| /* ── SECTION HEADERS ── */ | |
| .section-label { | |
| font-family: var(--mono); | |
| font-size: 0.68rem; | |
| color: var(--cyan); | |
| letter-spacing: 4px; | |
| text-transform: uppercase; | |
| margin-bottom: 0.75rem; | |
| } | |
| h2 { | |
| font-size: clamp(1.8rem, 4vw, 3rem); | |
| font-weight: 700; | |
| letter-spacing: -1px; | |
| margin-bottom: 0.5rem; | |
| line-height: 1.15; | |
| } | |
| .section-desc { | |
| color: var(--muted); | |
| font-size: 0.9rem; | |
| line-height: 1.7; | |
| max-width: 640px; | |
| margin-bottom: 3rem; | |
| } | |
| /* ── DIVIDER ── */ | |
| .divider { | |
| border: none; | |
| border-top: 1px solid var(--border); | |
| margin: 0; | |
| } | |
| /* ── DATASET CARDS ── */ | |
| .dataset-grid { | |
| display: grid; | |
| grid-template-columns: 1fr 1fr; | |
| gap: 1.5rem; | |
| margin-top: 2rem; | |
| } | |
| .card { | |
| background: var(--bg2); | |
| border: 1px solid var(--border); | |
| border-radius: 12px; | |
| padding: 2rem; | |
| position: relative; | |
| overflow: hidden; | |
| transition: border-color 0.3s, transform 0.3s; | |
| } | |
| .card:hover { | |
| border-color: rgba(0,212,255,0.4); | |
| transform: translateY(-2px); | |
| } | |
| .card::before { | |
| content: ''; | |
| position: absolute; | |
| top: 0; left: 0; right: 0; | |
| height: 2px; | |
| background: linear-gradient(90deg, var(--cyan), transparent); | |
| } | |
| .card-title { | |
| font-family: var(--mono); | |
| font-size: 0.78rem; | |
| color: var(--cyan); | |
| letter-spacing: 2px; | |
| text-transform: uppercase; | |
| margin-bottom: 1.2rem; | |
| } | |
| .card h3 { | |
| font-size: 1.5rem; | |
| font-weight: 700; | |
| margin-bottom: 0.5rem; | |
| } | |
| .card p { | |
| font-size: 0.85rem; | |
| color: var(--muted); | |
| line-height: 1.6; | |
| margin-bottom: 1.5rem; | |
| } | |
| .dataset-stat { | |
| display: flex; | |
| justify-content: space-between; | |
| align-items: center; | |
| padding: 0.5rem 0; | |
| border-top: 1px solid var(--border); | |
| font-size: 0.82rem; | |
| } | |
| .dataset-stat span:first-child { color: var(--muted); } | |
| .dataset-stat span:last-child { font-family: var(--mono); color: var(--cyan); } | |
| /* ── TABS ── */ | |
| .tabs { | |
| display: flex; | |
| gap: 0; | |
| border: 1px solid var(--border); | |
| border-radius: 8px; | |
| overflow: hidden; | |
| margin-bottom: 2rem; | |
| width: fit-content; | |
| } | |
| .tab-btn { | |
| background: transparent; | |
| border: none; | |
| color: var(--muted); | |
| font-family: var(--mono); | |
| font-size: 0.72rem; | |
| letter-spacing: 2px; | |
| text-transform: uppercase; | |
| padding: 0.7rem 1.5rem; | |
| cursor: pointer; | |
| transition: all 0.2s; | |
| border-right: 1px solid var(--border); | |
| } | |
| .tab-btn:last-child { border-right: none; } | |
| .tab-btn.active { | |
| background: rgba(0,212,255,0.1); | |
| color: var(--cyan); | |
| } | |
| .tab-btn:hover:not(.active) { | |
| background: rgba(0,212,255,0.05); | |
| color: var(--text); | |
| } | |
| .tab-content { display: none; } | |
| .tab-content.active { display: block; } | |
| /* ── CHART AREA ── */ | |
| .chart-wrap { | |
| background: var(--bg2); | |
| border: 1px solid var(--border); | |
| border-radius: 12px; | |
| padding: 2rem; | |
| margin-bottom: 2rem; | |
| } | |
| .chart-title { | |
| font-family: var(--mono); | |
| font-size: 0.72rem; | |
| color: var(--muted); | |
| letter-spacing: 2px; | |
| text-transform: uppercase; | |
| margin-bottom: 1.5rem; | |
| } | |
| .bar-chart { display: flex; flex-direction: column; gap: 0.65rem; } | |
| .bar-row { | |
| display: grid; | |
| grid-template-columns: 130px 1fr 70px; | |
| align-items: center; | |
| gap: 0.75rem; | |
| } | |
| .bar-label { | |
| font-family: var(--mono); | |
| font-size: 0.72rem; | |
| color: var(--muted); | |
| text-align: right; | |
| white-space: nowrap; | |
| } | |
| .bar-track { | |
| height: 20px; | |
| background: rgba(255,255,255,0.04); | |
| border-radius: 4px; | |
| overflow: hidden; | |
| } | |
| .bar-fill { | |
| height: 100%; | |
| border-radius: 4px; | |
| width: 0; | |
| transition: width 1s cubic-bezier(.16,1,.3,1); | |
| position: relative; | |
| } | |
| .bar-fill::after { | |
| content: ''; | |
| position: absolute; | |
| top: 0; right: 0; bottom: 0; | |
| width: 6px; | |
| background: rgba(255,255,255,0.3); | |
| border-radius: 0 4px 4px 0; | |
| } | |
| .bar-val { | |
| font-family: var(--mono); | |
| font-size: 0.72rem; | |
| color: var(--text); | |
| } | |
| /* ── DATA TABLE ── */ | |
| .table-wrap { | |
| overflow-x: auto; | |
| border: 1px solid var(--border); | |
| border-radius: 12px; | |
| } | |
| table { | |
| width: 100%; | |
| border-collapse: collapse; | |
| font-size: 0.8rem; | |
| } | |
| thead th { | |
| background: var(--bg3); | |
| font-family: var(--mono); | |
| font-size: 0.65rem; | |
| letter-spacing: 2px; | |
| text-transform: uppercase; | |
| color: var(--cyan); | |
| padding: 0.9rem 1rem; | |
| text-align: left; | |
| border-bottom: 1px solid var(--border); | |
| white-space: nowrap; | |
| } | |
| tbody td { | |
| padding: 0.7rem 1rem; | |
| border-bottom: 1px solid var(--border); | |
| color: var(--text); | |
| font-family: var(--mono); | |
| font-size: 0.78rem; | |
| white-space: nowrap; | |
| } | |
| tbody tr:last-child td { border-bottom: none; } | |
| tbody tr:hover td { background: rgba(0,212,255,0.04); } | |
| .best { color: var(--green); font-weight: 700; } | |
| .worst { color: var(--red); } | |
| .mid { color: var(--yellow); } | |
| .badge { | |
| display: inline-block; | |
| padding: 0.15rem 0.5rem; | |
| border-radius: 3px; | |
| font-size: 0.65rem; | |
| letter-spacing: 1px; | |
| text-transform: uppercase; | |
| } | |
| .badge-ml { background: rgba(0,255,157,0.1); color: var(--green); border: 1px solid rgba(0,255,157,0.2); } | |
| .badge-dl { background: rgba(0,212,255,0.1); color: var(--cyan); border: 1px solid rgba(0,212,255,0.2); } | |
| .badge-smote { background: rgba(255,107,53,0.1); color: var(--orange); border: 1px solid rgba(255,107,53,0.2); } | |
| /* ── FINDING CARDS ── */ | |
| .findings-grid { | |
| display: grid; | |
| grid-template-columns: repeat(3, 1fr); | |
| gap: 1.5rem; | |
| margin-top: 2rem; | |
| } | |
| .finding-card { | |
| background: var(--bg2); | |
| border: 1px solid var(--border); | |
| border-radius: 12px; | |
| padding: 1.75rem; | |
| position: relative; | |
| overflow: hidden; | |
| transition: all 0.3s; | |
| } | |
| .finding-card:hover { | |
| border-color: rgba(0,212,255,0.35); | |
| transform: translateY(-3px); | |
| } | |
| .finding-icon { | |
| font-size: 1.8rem; | |
| margin-bottom: 1rem; | |
| } | |
| .finding-card h4 { | |
| font-size: 1rem; | |
| font-weight: 700; | |
| margin-bottom: 0.6rem; | |
| color: var(--text); | |
| } | |
| .finding-card p { | |
| font-size: 0.82rem; | |
| color: var(--muted); | |
| line-height: 1.6; | |
| } | |
| /* ── COMPUTATIONAL COST ── */ | |
| .compute-grid { | |
| display: grid; | |
| grid-template-columns: 1fr 1fr; | |
| gap: 1.5rem; | |
| margin-top: 2rem; | |
| } | |
| /* ── GUIDELINES TABLE ── */ | |
| .guidelines-wrap { | |
| background: var(--bg2); | |
| border: 1px solid var(--border); | |
| border-radius: 12px; | |
| overflow: hidden; | |
| margin-top: 2rem; | |
| } | |
| .gl-row { | |
| display: grid; | |
| grid-template-columns: 2fr 1.2fr 1.2fr 2fr; | |
| gap: 0; | |
| border-bottom: 1px solid var(--border); | |
| } | |
| .gl-row:last-child { border-bottom: none; } | |
| .gl-header .gl-cell { | |
| background: var(--bg3); | |
| font-family: var(--mono); | |
| font-size: 0.65rem; | |
| color: var(--cyan); | |
| letter-spacing: 2px; | |
| text-transform: uppercase; | |
| } | |
| .gl-cell { | |
| padding: 1rem 1.2rem; | |
| font-size: 0.82rem; | |
| border-right: 1px solid var(--border); | |
| line-height: 1.5; | |
| } | |
| .gl-cell:last-child { border-right: none; } | |
| .gl-row:not(.gl-header):hover .gl-cell { background: rgba(0,212,255,0.03); } | |
| /* ── ANIMATIONS ── */ | |
| @keyframes fadeUp { | |
| from { opacity: 0; transform: translateY(24px); } | |
| to { opacity: 1; transform: translateY(0); } | |
| } | |
| .reveal { | |
| opacity: 0; | |
| transform: translateY(30px); | |
| transition: opacity 0.7s ease, transform 0.7s ease; | |
| } | |
| .reveal.visible { | |
| opacity: 1; | |
| transform: translateY(0); | |
| } | |
| /* ── FOOTER ── */ | |
| footer { | |
| position: relative; | |
| z-index: 1; | |
| border-top: 1px solid var(--border); | |
| padding: 2rem 4rem; | |
| display: flex; | |
| justify-content: space-between; | |
| align-items: center; | |
| } | |
| footer p { | |
| font-family: var(--mono); | |
| font-size: 0.68rem; | |
| color: var(--muted); | |
| letter-spacing: 1px; | |
| } | |
| .dot { | |
| display: inline-block; | |
| width: 6px; height: 6px; | |
| border-radius: 50%; | |
| background: var(--cyan); | |
| margin-right: 0.5rem; | |
| animation: blink 1.2s step-end infinite; | |
| } | |
| @keyframes blink { | |
| 0%, 100% { opacity: 1; } | |
| 50% { opacity: 0; } | |
| } | |
| @media (max-width: 768px) { | |
| #hero { padding: 6rem 2rem; } | |
| .hero-stats { flex-wrap: wrap; gap: 2rem; } | |
| .dataset-grid, .findings-grid, .compute-grid { grid-template-columns: 1fr; } | |
| .gl-row { grid-template-columns: 1fr 1fr; } | |
| nav { padding: 0 1rem; } | |
| .nav-links { display: none; } | |
| h1 { letter-spacing: -1px; } | |
| } | |
| </style> | |
| </head> | |
| <body> | |
| <div class="grid-bg"></div> | |
| <!-- NAV --> | |
| <nav> | |
| <div class="nav-logo"><span class="dot"></span>NIDS Survey 2024</div> | |
| <ul class="nav-links"> | |
| <li><a href="#datasets">Datasets</a></li> | |
| <li><a href="#results">Results</a></li> | |
| <li><a href="#compute">Compute Cost</a></li> | |
| <li><a href="#findings">Findings</a></li> | |
| <li><a href="#guidelines">Guidelines</a></li> | |
| </ul> | |
| </nav> | |
| <!-- HERO --> | |
| <div id="hero"> | |
| <div class="hero-orb"></div> | |
| <div class="hero-tag">Conference Presentation · NIDS Survey</div> | |
| <h1>From <span>Data</span><br>to Defense</h1> | |
| <p class="hero-sub">An analytical overview of machine and deep learning models for Network Intrusion Detection Systems, evaluated across two benchmark datasets under a unified experimental framework.</p> | |
| <div class="hero-stats"> | |
| <div class="stat-item"> | |
| <div class="stat-num">7</div> | |
| <div class="stat-label">Models Evaluated</div> | |
| </div> | |
| <div class="stat-item"> | |
| <div class="stat-num">2</div> | |
| <div class="stat-label">Benchmark Datasets</div> | |
| </div> | |
| <div class="stat-item"> | |
| <div class="stat-num">8</div> | |
| <div class="stat-label">Evaluation Metrics</div> | |
| </div> | |
| <div class="stat-item"> | |
| <div class="stat-num">4M+</div> | |
| <div class="stat-label">Records Processed</div> | |
| </div> | |
| </div> | |
| </div> | |
| <hr class="divider"> | |
| <!-- DATASETS --> | |
| <section id="datasets"> | |
| <div class="reveal"> | |
| <div class="section-label">// Section 01</div> | |
| <h2>Benchmark Datasets</h2> | |
| <p class="section-desc">Two widely-used intrusion detection datasets selected to represent both classical and modern network traffic environments.</p> | |
| </div> | |
| <div class="dataset-grid reveal"> | |
| <div class="card"> | |
| <div class="card-title">Dataset 01</div> | |
| <h3>KDD Cup 99</h3> | |
| <p>One of the earliest and most widely-used IDS benchmarks. Contains diverse attack types across 41 features, enabling systematic ML evaluation despite its age.</p> | |
| <div class="dataset-stat"><span>Total Records</span><span>~4,000,000</span></div> | |
| <div class="dataset-stat"><span>Features</span><span>41</span></div> | |
| <div class="dataset-stat"><span>Attack Types</span><span>10 (after filtering)</span></div> | |
| <div class="dataset-stat"><span>Largest Class</span><span>smurf · 2,807,886</span></div> | |
| <div class="dataset-stat"><span>Normal Traffic</span><span>972,781</span></div> | |
| </div> | |
| <div class="card"> | |
| <div class="card-title">Dataset 02</div> | |
| <h3>UNSW-NB15</h3> | |
| <p>A modern dataset with realistic traffic patterns and diverse attack categories. Reflects contemporary network environments with significant class imbalance.</p> | |
| <div class="dataset-stat"><span>Total Records</span><span>~2,367,624</span></div> | |
| <div class="dataset-stat"><span>Attack Categories</span><span>9</span></div> | |
| <div class="dataset-stat"><span>Benign Traffic</span><span>2,237,731</span></div> | |
| <div class="dataset-stat"><span>Rarest Class</span><span>Worms · 158</span></div> | |
| <div class="dataset-stat"><span>Imbalance Ratio</span><span>High</span></div> | |
| </div> | |
| </div> | |
| </section> | |
| <hr class="divider"> | |
| <!-- RESULTS --> | |
| <section id="results"> | |
| <div class="reveal"> | |
| <div class="section-label">// Section 02</div> | |
| <h2>Model Performance Results</h2> | |
| <p class="section-desc">Comprehensive metrics including balanced accuracy, F1-macro, G-Mean, precision, and recall evaluated across all models on both datasets — with and without SMOTE oversampling.</p> | |
| </div> | |
| <div class="reveal"> | |
| <div class="tabs"> | |
| <button class="tab-btn active" onclick="switchTab(event,'unsw-raw')">UNSW · Raw</button> | |
| <button class="tab-btn" onclick="switchTab(event,'unsw-smote')">UNSW · SMOTE</button> | |
| <button class="tab-btn" onclick="switchTab(event,'kdd-raw')">KDD Cup · Raw</button> | |
| <button class="tab-btn" onclick="switchTab(event,'kdd-smote')">KDD Cup · SMOTE</button> | |
| </div> | |
| <!-- UNSW RAW --> | |
| <div id="unsw-raw" class="tab-content active"> | |
| <div class="chart-wrap"> | |
| <div class="chart-title">// Balanced Accuracy — UNSW-NB15 (Raw)</div> | |
| <div class="bar-chart" id="chart-unsw-raw"></div> | |
| </div> | |
| <div class="table-wrap"> | |
| <table> | |
| <thead> | |
| <tr> | |
| <th>Model</th><th>Type</th><th>Accuracy</th><th>Balanced Acc.</th><th>Precision</th><th>Recall</th><th>F1-Macro</th><th>G-Mean</th> | |
| </tr> | |
| </thead> | |
| <tbody> | |
| <tr><td>Random Forest</td><td><span class="badge badge-ml">ML</span></td><td>0.9757</td><td class="best">0.5657</td><td class="best">0.6197</td><td class="best">0.5657</td><td class="best">0.5810</td><td class="best">0.4822</td></tr> | |
| <tr><td>XGBoost</td><td><span class="badge badge-ml">ML</span></td><td>0.9738</td><td>0.4247</td><td>0.6435</td><td>0.4247</td><td>0.4510</td><td class="worst">0.0180</td></tr> | |
| <tr><td>Decision Tree</td><td><span class="badge badge-ml">ML</span></td><td>0.9753</td><td>0.4703</td><td>0.5760</td><td>0.4703</td><td>0.4850</td><td>0.0238</td></tr> | |
| <tr><td>LSTM-CNN</td><td><span class="badge badge-dl">DL</span></td><td>0.9729</td><td>0.3823</td><td>0.4673</td><td>0.3823</td><td>0.3950</td><td>0.0001</td></tr> | |
| <tr><td>ANN</td><td><span class="badge badge-dl">DL</span></td><td>0.9589</td><td>0.3744</td><td>0.3568</td><td>0.3744</td><td>0.2990</td><td class="worst">≈0</td></tr> | |
| <tr><td>LSTM</td><td><span class="badge badge-dl">DL</span></td><td>0.9673</td><td>0.3287</td><td>0.3546</td><td>0.3287</td><td>0.3250</td><td class="worst">≈0</td></tr> | |
| <tr><td>CNN</td><td><span class="badge badge-dl">DL</span></td><td class="worst">0.9524</td><td class="worst">0.2781</td><td class="worst">0.3386</td><td class="worst">0.2781</td><td class="worst">0.2730</td><td class="worst">≈0</td></tr> | |
| </tbody> | |
| </table> | |
| </div> | |
| </div> | |
| <!-- UNSW SMOTE --> | |
| <div id="unsw-smote" class="tab-content"> | |
| <div class="chart-wrap"> | |
| <div class="chart-title">// Balanced Accuracy — UNSW-NB15 (After SMOTE) — Degradation observed</div> | |
| <div class="bar-chart" id="chart-unsw-smote"></div> | |
| </div> | |
| <div class="table-wrap"> | |
| <table> | |
| <thead> | |
| <tr> | |
| <th>Model</th><th>Accuracy</th><th>Balanced Acc.</th><th>Precision</th><th>Recall</th><th>F1-Macro</th><th>G-Mean</th> | |
| </tr> | |
| </thead> | |
| <tbody> | |
| <tr><td>LSTM-CNN <span class="badge badge-smote">SMOTE</span></td><td class="worst">0.0112</td><td>0.1241</td><td>0.0918</td><td>0.1241</td><td class="worst">0.0077</td><td>0.0017</td></tr> | |
| <tr><td>Random Forest <span class="badge badge-smote">SMOTE</span></td><td>0.9194</td><td class="best">0.1001</td><td class="best">0.1919</td><td class="best">0.1001</td><td class="best">0.0961</td><td class="worst">≈0</td></tr> | |
| <tr><td>ANN <span class="badge badge-smote">SMOTE</span></td><td>0.9194</td><td>0.1000</td><td>0.0919</td><td>0.1000</td><td>0.0958</td><td class="worst">≈0</td></tr> | |
| <tr><td>CNN <span class="badge badge-smote">SMOTE</span></td><td>0.9194</td><td>0.1000</td><td>0.0919</td><td>0.1000</td><td>0.0958</td><td class="worst">≈0</td></tr> | |
| <tr><td>XGBoost <span class="badge badge-smote">SMOTE</span></td><td>0.1104</td><td>0.1108</td><td>0.1044</td><td>0.1108</td><td>0.0248</td><td class="worst">≈0</td></tr> | |
| <tr><td>Decision Tree <span class="badge badge-smote">SMOTE</span></td><td>0.8633</td><td>0.0960</td><td>0.0930</td><td>0.0960</td><td>0.0937</td><td class="worst">≈0</td></tr> | |
| <tr><td>LSTM <span class="badge badge-smote">SMOTE</span></td><td class="worst">0.0352</td><td class="worst">0.0683</td><td class="worst">0.0814</td><td class="worst">0.0683</td><td class="worst">0.0097</td><td class="worst">≈0</td></tr> | |
| </tbody> | |
| </table> | |
| </div> | |
| <div class="chart-wrap" style="margin-top:1.5rem; border-color: rgba(255,61,90,0.2)"> | |
| <div class="chart-title" style="color:var(--red)">// ⚠ SMOTE degraded ALL metrics on UNSW-NB15 — Not recommended for high-dimensional imbalanced data</div> | |
| <p style="font-size:0.85rem; color:var(--muted); line-height:1.7">Applying SMOTE to the UNSW-NB15 dataset caused substantial deterioration across all models. Balanced accuracy, F1-macro, and G-Mean dropped significantly compared to raw data settings, indicating that sample-level oversampling is ineffective for high-dimensional complex datasets. Model robustness and algorithm-level imbalance handling are more critical.</p> | |
| </div> | |
| </div> | |
| <!-- KDD RAW --> | |
| <div id="kdd-raw" class="tab-content"> | |
| <div class="chart-wrap"> | |
| <div class="chart-title">// Balanced Accuracy — KDD Cup 99 (Raw)</div> | |
| <div class="bar-chart" id="chart-kdd-raw"></div> | |
| </div> | |
| <div class="table-wrap"> | |
| <table> | |
| <thead> | |
| <tr> | |
| <th>Model</th><th>Type</th><th>Accuracy</th><th>Balanced Acc.</th><th>Precision</th><th>Recall</th><th>F1-Macro</th><th>G-Mean</th> | |
| </tr> | |
| </thead> | |
| <tbody> | |
| <tr><td>Random Forest</td><td><span class="badge badge-ml">ML</span></td><td class="best">0.9999</td><td class="best">0.9940</td><td class="best">0.9969</td><td class="best">0.9940</td><td class="best">0.9955</td><td class="best">0.9939</td></tr> | |
| <tr><td>XGBoost</td><td><span class="badge badge-ml">ML</span></td><td>0.9997</td><td>0.9909</td><td>0.9941</td><td>0.9909</td><td>0.9925</td><td>0.9908</td></tr> | |
| <tr><td>LSTM-CNN</td><td><span class="badge badge-dl">DL</span></td><td>0.9992</td><td>0.9848</td><td>0.9687</td><td>0.9848</td><td>0.9758</td><td>0.9846</td></tr> | |
| <tr><td>Decision Tree</td><td><span class="badge badge-ml">ML</span></td><td>0.9988</td><td>0.9614</td><td>0.9659</td><td>0.9614</td><td>0.9622</td><td>0.9586</td></tr> | |
| <tr><td>LSTM</td><td><span class="badge badge-dl">DL</span></td><td>0.9985</td><td>0.9673</td><td>0.9401</td><td>0.9673</td><td>0.9517</td><td>0.9663</td></tr> | |
| <tr><td>ANN</td><td><span class="badge badge-dl">DL</span></td><td>0.9987</td><td>0.9427</td><td>0.9622</td><td>0.9427</td><td>0.9519</td><td>0.9362</td></tr> | |
| <tr><td>CNN</td><td><span class="badge badge-dl">DL</span></td><td>0.9899</td><td class="worst">0.6256</td><td>0.7080</td><td class="worst">0.6256</td><td>0.6446</td><td class="worst">0.0051</td></tr> | |
| </tbody> | |
| </table> | |
| </div> | |
| </div> | |
| <!-- KDD SMOTE --> | |
| <div id="kdd-smote" class="tab-content"> | |
| <div class="chart-wrap"> | |
| <div class="chart-title">// Balanced Accuracy — KDD Cup 99 (After SMOTE) — Selective improvements</div> | |
| <div class="bar-chart" id="chart-kdd-smote"></div> | |
| </div> | |
| <div class="table-wrap"> | |
| <table> | |
| <thead> | |
| <tr> | |
| <th>Model</th><th>Accuracy</th><th>Balanced Acc.</th><th>Precision</th><th>Recall</th><th>F1-Macro</th><th>G-Mean</th> | |
| </tr> | |
| </thead> | |
| <tbody> | |
| <tr><td>Random Forest <span class="badge badge-smote">SMOTE</span></td><td class="best">0.9999</td><td>0.9922</td><td class="best">0.9978</td><td>0.9922</td><td class="best">0.9949</td><td>0.9920</td></tr> | |
| <tr><td>XGBoost <span class="badge badge-smote">SMOTE</span></td><td>0.9997</td><td class="best">0.9971</td><td>0.9864</td><td class="best">0.9971</td><td>0.9916</td><td class="best">0.9971</td></tr> | |
| <tr><td>ANN <span class="badge badge-smote">SMOTE</span></td><td>0.9926</td><td>0.9953</td><td>0.7993</td><td>0.9954</td><td>0.8655</td><td>0.9953</td></tr> | |
| <tr><td>LSTM-CNN <span class="badge badge-smote">SMOTE</span></td><td>0.9992</td><td>0.9907</td><td>0.9589</td><td>0.9908</td><td>0.9737</td><td>0.9906</td></tr> | |
| <tr><td>Decision Tree <span class="badge badge-smote">SMOTE</span></td><td>0.9970</td><td>0.9931</td><td>0.9252</td><td>0.9931</td><td>0.9386</td><td>0.9930</td></tr> | |
| <tr><td>LSTM <span class="badge badge-smote">SMOTE</span></td><td>0.9981</td><td>0.9980</td><td>0.9111</td><td>0.9912</td><td>0.9431</td><td>0.9910</td></tr> | |
| <tr><td>CNN <span class="badge badge-smote">SMOTE</span></td><td>0.9494</td><td>0.9721</td><td class="worst">0.5231</td><td>0.9721</td><td class="worst">0.6199</td><td>0.9717</td></tr> | |
| </tbody> | |
| </table> | |
| </div> | |
| <div class="chart-wrap" style="margin-top:1.5rem; border-color: rgba(0,255,157,0.2)"> | |
| <div class="chart-title" style="color:var(--green)">// ✓ SMOTE on KDD Cup 99 — Improvements for DL models, marginal effect on tree-based</div> | |
| <p style="font-size:0.85rem; color:var(--muted); line-height:1.7">SMOTE improved minority class recall for deep learning models notably (ANN: 0.94 → 0.99, CNN: 0.63 → 0.97). However, precision dropped, indicating more false positives. Tree-based models (RF, XGBoost) were largely unaffected — they inherently handle class imbalance well and do not benefit significantly from data-level oversampling.</p> | |
| </div> | |
| </div> | |
| </div> | |
| </section> | |
| <hr class="divider"> | |
| <!-- COMPUTE COST --> | |
| <section id="compute"> | |
| <div class="reveal"> | |
| <div class="section-label">// Section 03</div> | |
| <h2>Computational Cost Analysis</h2> | |
| <p class="section-desc">Inference time and memory footprint are critical factors for real-world NIDS deployment. Deep learning architectures show significantly higher latency and resource consumption.</p> | |
| </div> | |
| <div class="compute-grid reveal"> | |
| <div class="chart-wrap"> | |
| <div class="chart-title">// Inference Time (seconds) — UNSW-NB15</div> | |
| <div class="bar-chart" id="chart-time-unsw"></div> | |
| </div> | |
| <div class="chart-wrap"> | |
| <div class="chart-title">// Inference Time (seconds) — KDD Cup 99</div> | |
| <div class="bar-chart" id="chart-time-kdd"></div> | |
| </div> | |
| </div> | |
| <div class="compute-grid reveal"> | |
| <div class="chart-wrap"> | |
| <div class="chart-title">// Memory Usage (MB) — UNSW-NB15 · Note: RF 22,424 MB</div> | |
| <div class="bar-chart" id="chart-mem-unsw"></div> | |
| </div> | |
| <div class="chart-wrap"> | |
| <div class="chart-title">// Memory Usage (MB) — KDD Cup 99</div> | |
| <div class="bar-chart" id="chart-mem-kdd"></div> | |
| </div> | |
| </div> | |
| <div class="table-wrap reveal" style="margin-top:1.5rem"> | |
| <table> | |
| <thead> | |
| <tr><th>Model</th><th>Type</th><th>UNSW Infer. Time (s)</th><th>UNSW Memory (MB)</th><th>KDD Infer. Time (s)</th><th>KDD Memory (MB)</th></tr> | |
| </thead> | |
| <tbody> | |
| <tr><td>Decision Tree</td><td><span class="badge badge-ml">ML</span></td><td class="best">0.025</td><td>1,579</td><td class="best">0.026</td><td>4,959</td></tr> | |
| <tr><td>XGBoost</td><td><span class="badge badge-ml">ML</span></td><td>0.333</td><td class="best">1,571</td><td>0.226</td><td class="best">4,959</td></tr> | |
| <tr><td>Random Forest</td><td><span class="badge badge-ml">ML</span></td><td>0.571</td><td class="worst">22,424</td><td>0.373</td><td>4,989</td></tr> | |
| <tr><td>ANN</td><td><span class="badge badge-dl">DL</span></td><td>1.622</td><td>5,267</td><td>1.410</td><td>4,503</td></tr> | |
| <tr><td>CNN</td><td><span class="badge badge-dl">DL</span></td><td>1.649</td><td>4,375</td><td>1.314</td><td>5,096</td></tr> | |
| <tr><td>LSTM</td><td><span class="badge badge-dl">DL</span></td><td>1.902</td><td>4,710</td><td>1.475</td><td>5,104</td></tr> | |
| <tr><td>LSTM-CNN</td><td><span class="badge badge-dl">DL</span></td><td class="worst">3.721</td><td>4,720</td><td class="worst">2.995</td><td>5,112</td></tr> | |
| </tbody> | |
| </table> | |
| </div> | |
| </section> | |
| <hr class="divider"> | |
| <!-- KEY FINDINGS --> | |
| <section id="findings"> | |
| <div class="reveal"> | |
| <div class="section-label">// Section 04</div> | |
| <h2>Key Findings</h2> | |
| <p class="section-desc">Six critical insights derived from unified experimental evaluation across models and datasets.</p> | |
| </div> | |
| <div class="findings-grid"> | |
| <div class="finding-card reveal"> | |
| <div class="finding-icon">🌲</div> | |
| <h4>Tree Models Dominate on Structured Data</h4> | |
| <p>Random Forest and XGBoost achieved near-perfect F1-macro (>0.99) on KDD Cup 99. Ensemble methods consistently outperformed all other approaches on structured, tabular datasets.</p> | |
| </div> | |
| <div class="finding-card reveal"> | |
| <div class="finding-icon">⚖️</div> | |
| <h4>Accuracy is a Misleading Metric</h4> | |
| <p>Multiple DL models reported >96% accuracy on UNSW-NB15, yet their balanced accuracy and G-Mean revealed near-zero minority class detection, exposing the danger of relying on a single metric.</p> | |
| </div> | |
| <div class="finding-card reveal"> | |
| <div class="finding-icon">🧠</div> | |
| <h4>Deep Learning: High Cost, Mixed Returns</h4> | |
| <p>CNN-LSTM hybrid achieved strong KDD performance (F1: 0.976) but requires 3.7s inference and >5GB memory. The computational cost limits real-time applicability significantly.</p> | |
| </div> | |
| <div class="finding-card reveal"> | |
| <div class="finding-icon">🚫</div> | |
| <h4>SMOTE Harmful on High-Dimensional Data</h4> | |
| <p>Applying SMOTE to UNSW-NB15 degraded all metrics across all models substantially. Simple sample-level balancing cannot compensate for feature complexity and distribution shifts.</p> | |
| </div> | |
| <div class="finding-card reveal"> | |
| <div class="finding-icon">✅</div> | |
| <h4>SMOTE Helps DL on KDD Cup 99</h4> | |
| <p>ANN balanced accuracy improved from 0.943 to 0.995, and CNN from 0.626 to 0.972 after SMOTE on KDD Cup 99. However, this came at the cost of precision degradation.</p> | |
| </div> | |
| <div class="finding-card reveal"> | |
| <div class="finding-icon">⚡</div> | |
| <h4>Decision Tree Best for Real-Time</h4> | |
| <p>With ~0.025s inference time and the smallest memory footprint, Decision Trees are optimal for latency-sensitive deployments where sub-second detection is a hard requirement.</p> | |
| </div> | |
| </div> | |
| </section> | |
| <hr class="divider"> | |
| <!-- GUIDELINES --> | |
| <section id="guidelines"> | |
| <div class="reveal"> | |
| <div class="section-label">// Section 05</div> | |
| <h2>Dataset–Model Selection Guidelines</h2> | |
| <p class="section-desc">Practical decision framework for selecting the right model and strategy based on deployment requirements and dataset characteristics.</p> | |
| </div> | |
| <div class="guidelines-wrap reveal"> | |
| <div class="gl-row gl-header"> | |
| <div class="gl-cell">Deployment Scenario</div> | |
| <div class="gl-cell">Recommended Model</div> | |
| <div class="gl-cell">Dataset</div> | |
| <div class="gl-cell">Key Evidence</div> | |
| </div> | |
| <div class="gl-row"> | |
| <div class="gl-cell">Structured, low-complexity traffic</div> | |
| <div class="gl-cell"><span class="badge badge-ml">RF / XGBoost</span></div> | |
| <div class="gl-cell" style="font-family:var(--mono); font-size:0.78rem; color:var(--cyan)">KDD Cup 99</div> | |
| <div class="gl-cell" style="color:var(--muted)">F1-macro >0.99; near-perfect balanced accuracy</div> | |
| </div> | |
| <div class="gl-row"> | |
| <div class="gl-cell">High-dimensional, imbalanced traffic</div> | |
| <div class="gl-cell"><span class="badge badge-ml">RF</span></div> | |
| <div class="gl-cell" style="font-family:var(--mono); font-size:0.78rem; color:var(--cyan)">UNSW-NB15</div> | |
| <div class="gl-cell" style="color:var(--muted)">Best balanced accuracy (0.57) and G-Mean (0.48) among all tested models</div> | |
| </div> | |
| <div class="gl-row"> | |
| <div class="gl-cell">Real-time / latency-sensitive deployment</div> | |
| <div class="gl-cell"><span class="badge badge-ml">DT</span></div> | |
| <div class="gl-cell" style="font-family:var(--mono); font-size:0.78rem; color:var(--cyan)">Both</div> | |
| <div class="gl-cell" style="color:var(--muted)">Fastest inference (~0.02s); lowest memory footprint across all datasets</div> | |
| </div> | |
| <div class="gl-row"> | |
| <div class="gl-cell">Sequential / temporal attack patterns</div> | |
| <div class="gl-cell"><span class="badge badge-dl">LSTM / CNN-LSTM</span></div> | |
| <div class="gl-cell" style="font-family:var(--mono); font-size:0.78rem; color:var(--cyan)">KDD Cup 99</div> | |
| <div class="gl-cell" style="color:var(--muted)">Strong recall on ordered flow attacks; temporal dependency modeling</div> | |
| </div> | |
| <div class="gl-row"> | |
| <div class="gl-cell">Imbalanced data + deep learning</div> | |
| <div class="gl-cell"><span class="badge badge-smote">SMOTE + ANN/LSTM</span></div> | |
| <div class="gl-cell" style="font-family:var(--mono); font-size:0.78rem; color:var(--cyan)">KDD Cup 99</div> | |
| <div class="gl-cell" style="color:var(--muted)">Balanced accuracy improved from 0.94 to 0.99; recall gains outweigh precision drop</div> | |
| </div> | |
| <div class="gl-row"> | |
| <div class="gl-cell">High-dimensional data + imbalance</div> | |
| <div class="gl-cell"><span class="badge badge-ml">RF (no SMOTE)</span></div> | |
| <div class="gl-cell" style="font-family:var(--mono); font-size:0.78rem; color:var(--cyan)">UNSW-NB15</div> | |
| <div class="gl-cell" style="color:var(--muted)">SMOTE degraded all metrics on high-dimensional data; avoid sample-level balancing</div> | |
| </div> | |
| </div> | |
| <!-- Conclusion box --> | |
| <div class="card reveal" style="margin-top:2rem; border-color:rgba(0,255,157,0.2)"> | |
| <div class="card-title" style="color:var(--green)">// Conclusion & Future Directions</div> | |
| <h3 style="margin-bottom:1rem; font-size:1.2rem">Ensemble Methods Offer the Best Production Trade-off</h3> | |
| <p>Decision Trees achieve inference times as low as 0.02s for real-time monitoring, while hybrid CNN-LSTM models exceed 3s latency with >5GB memory — suitable only for offline analysis. RF and XGBoost provide the optimal balance of strong detection performance with manageable inference costs, making them the most practical choice for production NIDS environments.</p> | |
| <p style="margin-top:1rem">Future research should explore <strong style="color:var(--cyan)">transformer-based architectures</strong> for NIDS — leveraging self-attention for parallel processing and better capture of global traffic patterns, potentially overcoming the computational bottlenecks observed in LSTM and CNN-LSTM models.</p> | |
| </div> | |
| </section> | |
| <footer> | |
| <p>From Data to Defense · NIDS Survey · Evaluation of ML & DL for Network Intrusion Detection</p> | |
| <p style="color:var(--cyan)">KDD Cup 99 · UNSW-NB15 · 7 Models · Unified Evaluation Framework</p> | |
| </footer> | |
| <script> | |
| // ── BAR CHART RENDERER ── | |
| function renderBar(containerId, data, colorFn) { | |
| const el = document.getElementById(containerId); | |
| if (!el) return; | |
| const max = Math.max(...data.map(d => d.val)); | |
| el.innerHTML = ''; | |
| data.forEach((d, i) => { | |
| const pct = (d.val / max * 100).toFixed(1); | |
| const color = colorFn ? colorFn(d, i) : `hsl(${180 + i * 15}, 70%, 55%)`; | |
| const row = document.createElement('div'); | |
| row.className = 'bar-row'; | |
| row.innerHTML = ` | |
| <div class="bar-label">${d.label}</div> | |
| <div class="bar-track"><div class="bar-fill" data-width="${pct}" style="background:${color}; width:0%"></div></div> | |
| <div class="bar-val">${d.val.toFixed ? d.val.toFixed(4) : d.val}</div>`; | |
| el.appendChild(row); | |
| }); | |
| requestAnimationFrame(() => { | |
| el.querySelectorAll('.bar-fill').forEach(b => b.style.width = b.dataset.width + '%'); | |
| }); | |
| } | |
| const ml_color = () => 'linear-gradient(90deg, #00d4ff, #00a8cc)'; | |
| const dl_color = () => 'linear-gradient(90deg, #8b5cf6, #6d28d9)'; | |
| const smote_color = () => 'linear-gradient(90deg, #ff6b35, #cc4400)'; | |
| function modelColor(d) { | |
| if (['RF','DT','XGBoost'].includes(d.type)) return 'linear-gradient(90deg, #00d4ff, #00a8cc)'; | |
| return 'linear-gradient(90deg, #8b5cf6, #6d28d9)'; | |
| } | |
| // UNSW RAW balanced acc | |
| renderBar('chart-unsw-raw', [ | |
| {label:'Random Forest', val:0.5657, type:'RF'}, | |
| {label:'Decision Tree', val:0.4703, type:'DT'}, | |
| {label:'XGBoost', val:0.4247, type:'XGBoost'}, | |
| {label:'LSTM-CNN', val:0.3823, type:'dl'}, | |
| {label:'ANN', val:0.3744, type:'dl'}, | |
| {label:'LSTM', val:0.3287, type:'dl'}, | |
| {label:'CNN', val:0.2781, type:'dl'}, | |
| ], modelColor); | |
| // UNSW SMOTE balanced acc | |
| renderBar('chart-unsw-smote', [ | |
| {label:'LSTM-CNN', val:0.1241, type:'dl'}, | |
| {label:'XGBoost', val:0.1108, type:'XGBoost'}, | |
| {label:'Random Forest', val:0.1001, type:'RF'}, | |
| {label:'ANN', val:0.1000, type:'dl'}, | |
| {label:'CNN', val:0.1000, type:'dl'}, | |
| {label:'DT', val:0.0960, type:'DT'}, | |
| {label:'LSTM', val:0.0683, type:'dl'}, | |
| ], (d) => smote_color()); | |
| // KDD RAW balanced acc | |
| renderBar('chart-kdd-raw', [ | |
| {label:'Random Forest', val:0.9940, type:'RF'}, | |
| {label:'XGBoost', val:0.9909, type:'XGBoost'}, | |
| {label:'LSTM-CNN', val:0.9848, type:'dl'}, | |
| {label:'LSTM', val:0.9673, type:'dl'}, | |
| {label:'Decision Tree', val:0.9614, type:'DT'}, | |
| {label:'ANN', val:0.9427, type:'dl'}, | |
| {label:'CNN', val:0.6256, type:'dl'}, | |
| ], modelColor); | |
| // KDD SMOTE balanced acc | |
| renderBar('chart-kdd-smote', [ | |
| {label:'LSTM', val:0.9980, type:'dl'}, | |
| {label:'ANN', val:0.9953, type:'dl'}, | |
| {label:'XGBoost', val:0.9971, type:'XGBoost'}, | |
| {label:'DT', val:0.9931, type:'DT'}, | |
| {label:'Random Forest', val:0.9922, type:'RF'}, | |
| {label:'LSTM-CNN', val:0.9907, type:'dl'}, | |
| {label:'CNN', val:0.9721, type:'dl'}, | |
| ], (d) => smote_color()); | |
| // INFERENCE TIME UNSW | |
| renderBar('chart-time-unsw', [ | |
| {label:'DT', val:0.025, type:'DT'}, | |
| {label:'XGBoost', val:0.333, type:'XGBoost'}, | |
| {label:'RF', val:0.571, type:'RF'}, | |
| {label:'ANN', val:1.622, type:'dl'}, | |
| {label:'CNN', val:1.649, type:'dl'}, | |
| {label:'LSTM', val:1.902, type:'dl'}, | |
| {label:'LSTM-CNN', val:3.721, type:'dl'}, | |
| ], (d) => d.type === 'dl' ? 'linear-gradient(90deg,#ff3d5a,#cc2a45)' : 'linear-gradient(90deg,#00d4ff,#00a8cc)'); | |
| // INFERENCE TIME KDD | |
| renderBar('chart-time-kdd', [ | |
| {label:'DT', val:0.026, type:'DT'}, | |
| {label:'XGBoost', val:0.226, type:'XGBoost'}, | |
| {label:'RF', val:0.373, type:'RF'}, | |
| {label:'CNN', val:1.314, type:'dl'}, | |
| {label:'ANN', val:1.410, type:'dl'}, | |
| {label:'LSTM', val:1.475, type:'dl'}, | |
| {label:'LSTM-CNN', val:2.995, type:'dl'}, | |
| ], (d) => d.type === 'dl' ? 'linear-gradient(90deg,#ff3d5a,#cc2a45)' : 'linear-gradient(90deg,#00d4ff,#00a8cc)'); | |
| // MEMORY UNSW (cap RF for display) | |
| renderBar('chart-mem-unsw', [ | |
| {label:'XGBoost', val:1571, type:'XGBoost'}, | |
| {label:'DT', val:1580, type:'DT'}, | |
| {label:'CNN', val:4375, type:'dl'}, | |
| {label:'LSTM', val:4710, type:'dl'}, | |
| {label:'LSTM-CNN', val:4720, type:'dl'}, | |
| {label:'ANN', val:5267, type:'dl'}, | |
| {label:'RF ⚠', val:22424, type:'RF'}, | |
| ], (d) => d.label.includes('RF') ? 'linear-gradient(90deg,#ff6b35,#cc4400)' : d.type === 'dl' ? 'linear-gradient(90deg,#8b5cf6,#6d28d9)' : 'linear-gradient(90deg,#00d4ff,#00a8cc)'); | |
| // MEMORY KDD | |
| renderBar('chart-mem-kdd', [ | |
| {label:'ANN', val:4503, type:'dl'}, | |
| {label:'DT', val:4959, type:'DT'}, | |
| {label:'XGBoost', val:4959, type:'XGBoost'}, | |
| {label:'RF', val:4989, type:'RF'}, | |
| {label:'CNN', val:5096, type:'dl'}, | |
| {label:'LSTM', val:5104, type:'dl'}, | |
| {label:'LSTM-CNN', val:5112, type:'dl'}, | |
| ], (d) => d.type === 'dl' ? 'linear-gradient(90deg,#8b5cf6,#6d28d9)' : 'linear-gradient(90deg,#00d4ff,#00a8cc)'); | |
| // ── TAB SWITCH ── | |
| function switchTab(e, id) { | |
| document.querySelectorAll('.tab-btn').forEach(b => b.classList.remove('active')); | |
| document.querySelectorAll('.tab-content').forEach(c => c.classList.remove('active')); | |
| e.target.classList.add('active'); | |
| document.getElementById(id).classList.add('active'); | |
| // Re-trigger bar animations | |
| setTimeout(() => { | |
| document.querySelectorAll('#' + id + ' .bar-fill').forEach(b => { | |
| const w = b.dataset.width; | |
| b.style.width = '0%'; | |
| requestAnimationFrame(() => setTimeout(() => b.style.width = w + '%', 50)); | |
| }); | |
| }, 50); | |
| } | |
| // ── SCROLL REVEAL ── | |
| const observer = new IntersectionObserver((entries) => { | |
| entries.forEach(e => { if (e.isIntersecting) { e.target.classList.add('visible'); } }); | |
| }, { threshold: 0.1 }); | |
| document.querySelectorAll('.reveal').forEach(el => observer.observe(el)); | |
| // ── ANIMATE BARS ON LOAD ── | |
| setTimeout(() => { | |
| document.querySelectorAll('#unsw-raw .bar-fill').forEach(b => b.style.width = b.dataset.width + '%'); | |
| }, 300); | |
| </script> | |
| </body> | |
| </html> | |