RSK World - Named Entity Recognition Dataset - Project Files Browser | RSK World

<!DOCTYPE html>
<html lang="en">
<head>
    <!--
    Project: Named Entity Recognition Dataset
    Author: RSK World
    Website: https://rskworld.in
    Email: help@rskworld.in
    Phone: +91 93305 39277
    Description: Interactive demo page for Named Entity Recognition Dataset
    -->
    <meta charset="UTF-8">
    <meta name="viewport" content="width=device-width, initial-scale=1.0">
    <title>Named Entity Recognition Dataset - Demo | RSK World</title>
    <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.0/css/all.min.css">
    <style>
        * {
            margin: 0;
            padding: 0;
            box-sizing: border-box;
        }

        body {
            font-family: 'Segoe UI', Tahoma, Geneva, Verdana, sans-serif;
            background: linear-gradient(135deg, #667eea 0%, #764ba2 100%);
            min-height: 100vh;
            padding: 0;
            margin: 0;
            display: flex;
            flex-direction: column;
        }

        /* Navigation Header */
        header {
            background: linear-gradient(135deg, #667eea 0%, #764ba2 100%);
            color: white;
            padding: 0;
            box-shadow: 0 2px 10px rgba(0,0,0,0.1);
            position: sticky;
            top: 0;
            z-index: 1000;
            width: 100%;
            display: block;
            visibility: visible;
        }

        .header-top {
            background: rgba(0,0,0,0.2);
            padding: 10px 0;
            font-size: 14px;
            display: block;
            visibility: visible;
            width: 100%;
        }

        .header-top .container,
        .header-main .container {
            max-width: 1200px;
            margin: 0 auto;
            padding: 0 20px;
            display: flex;
            justify-content: space-between;
            align-items: center;
            flex-wrap: wrap;
            background: transparent;
            box-shadow: none;
            border-radius: 0;
            overflow: visible;
        }

        .header-top a {
            color: white;
            text-decoration: none;
            margin: 0 10px;
            transition: all 0.3s;
            display: inline-block;
        }

        .header-top a:hover {
            opacity: 0.8;
            transform: translateY(-2px);
        }

        .header-main {
            padding: 30px 20px;
            display: block;
            visibility: visible;
            width: 100%;
        }


        .logo {
            display: flex;
            align-items: center;
            gap: 15px;
        }

        .logo i {
            font-size: 2.5em;
            transition: transform 0.3s;
        }

        .logo:hover i {
            transform: rotate(360deg);
        }

        .logo-text h1 {
            font-size: 2em;
            margin-bottom: 5px;
        }

        .logo-text p {
            font-size: 0.9em;
            opacity: 0.9;
        }

        nav ul {
            list-style: none;
            display: flex;
            gap: 20px;
            flex-wrap: wrap;
        }

        nav a {
            color: white;
            text-decoration: none;
            padding: 10px 20px;
            border-radius: 5px;
            transition: all 0.3s;
            font-weight: 500;
            position: relative;
        }

        nav a::after {
            content: '';
            position: absolute;
            bottom: 5px;
            left: 50%;
            width: 0;
            height: 2px;
            background: white;
            transition: all 0.3s;
            transform: translateX(-50%);
        }

        nav a:hover::after {
            width: 80%;
        }

        nav a:hover {
            background: rgba(255,255,255,0.2);
            transform: translateY(-2px);
        }

        .main-container {
            max-width: 1200px;
            margin: 20px auto;
            background: white;
            border-radius: 15px;
            box-shadow: 0 10px 40px rgba(0,0,0,0.2);
            overflow: hidden;
        }
        
        /* Keep .container for backward compatibility but override for main content */
        .container:not(header .container):not(footer .container) {
            max-width: 1200px;
            margin: 20px auto;
            background: white;
            border-radius: 15px;
            box-shadow: 0 10px 40px rgba(0,0,0,0.2);
            overflow: hidden;
        }

        /* Ensure header containers don't inherit main container styles */
        header .container,
        header .header-top .container,
        header .header-main .container {
            background: transparent !important;
            box-shadow: none !important;
            border-radius: 0 !important;
            overflow: visible !important;
            margin: 0 !important;
            padding: 0 20px !important;
        }
        
        /* Ensure footer container doesn't inherit main container styles */
        footer .container {
            background: transparent !important;
            box-shadow: none !important;
            border-radius: 0 !important;
            overflow: visible !important;
            margin: 0 auto !important;
        }

        .page-header {
            background: linear-gradient(135deg, #667eea 0%, #764ba2 100%);
            color: white;
            padding: 40px;
            text-align: center;
        }

        .page-header h1 {
            font-size: 2.5em;
            margin-bottom: 10px;
        }

        .page-header p {
            font-size: 1.2em;
            opacity: 0.9;
        }

        .content {
            padding: 40px;
        }

        .section {
            margin-bottom: 40px;
        }

        .section h2 {
            color: #667eea;
            margin-bottom: 20px;
            font-size: 1.8em;
            border-bottom: 3px solid #667eea;
            padding-bottom: 10px;
        }

        .input-section {
            background: #f8f9fa;
            padding: 30px;
            border-radius: 10px;
            margin-bottom: 30px;
        }

        .input-group {
            margin-bottom: 20px;
        }

        label {
            display: block;
            margin-bottom: 8px;
            font-weight: 600;
            color: #333;
        }

        textarea {
            width: 100%;
            padding: 15px;
            border: 2px solid #ddd;
            border-radius: 8px;
            font-size: 16px;
            font-family: inherit;
            resize: vertical;
            min-height: 120px;
        }

        textarea:focus {
            outline: none;
            border-color: #667eea;
        }

        button {
            background: linear-gradient(135deg, #667eea 0%, #764ba2 100%);
            color: white;
            border: none;
            padding: 15px 30px;
            font-size: 16px;
            border-radius: 8px;
            cursor: pointer;
            font-weight: 600;
            transition: transform 0.2s, box-shadow 0.2s;
        }

        button:hover {
            transform: translateY(-2px);
            box-shadow: 0 5px 15px rgba(102, 126, 234, 0.4);
        }

        button:active {
            transform: translateY(0);
        }

        .result-section {
            margin-top: 30px;
        }

        .highlighted-text {
            background: white;
            padding: 20px;
            border-radius: 8px;
            border: 2px solid #ddd;
            line-height: 1.8;
            font-size: 16px;
            min-height: 100px;
        }

        .entity {
            padding: 3px 6px;
            border-radius: 4px;
            font-weight: 600;
            margin: 0 2px;
            display: inline-block;
        }

        .entity-PERSON {
            background-color: #FF6B6B;
            color: white;
        }

        .entity-ORG {
            background-color: #4ECDC4;
            color: white;
        }

        .entity-LOC {
            background-color: #45B7D1;
            color: white;
        }

        .entity-DATE {
            background-color: #FFA07A;
            color: white;
        }

        .entity-MONEY {
            background-color: #98D8C8;
            color: #333;
        }

        .entity-PERCENT {
            background-color: #F7DC6F;
            color: #333;
        }

        .legend {
            background: #f8f9fa;
            padding: 20px;
            border-radius: 8px;
            margin-top: 20px;
        }

        .legend h3 {
            margin-bottom: 15px;
            color: #333;
        }

        .legend-items {
            display: flex;
            flex-wrap: wrap;
            gap: 15px;
        }

        .legend-item {
            display: flex;
            align-items: center;
            gap: 8px;
        }

        .legend-color {
            width: 30px;
            height: 30px;
            border-radius: 4px;
        }

        .stats {
            display: grid;
            grid-template-columns: repeat(auto-fit, minmax(200px, 1fr));
            gap: 20px;
            margin-top: 20px;
        }

        .stat-card {
            background: linear-gradient(135deg, #667eea 0%, #764ba2 100%);
            color: white;
            padding: 20px;
            border-radius: 10px;
            text-align: center;
        }

        .stat-card h3 {
            font-size: 2em;
            margin-bottom: 5px;
        }

        .stat-card p {
            opacity: 0.9;
        }

        .sample-texts {
            display: grid;
            grid-template-columns: repeat(auto-fit, minmax(300px, 1fr));
            gap: 15px;
            margin-top: 20px;
        }

        .sample-card {
            background: #f8f9fa;
            padding: 15px;
            border-radius: 8px;
            cursor: pointer;
            transition: all 0.3s;
            border: 2px solid transparent;
        }

        .sample-card:hover {
            border-color: #667eea;
            transform: translateY(-3px);
            box-shadow: 0 5px 15px rgba(0,0,0,0.1);
        }

        /* Footer */
        footer {
            background: #2c3e50;
            color: white;
            padding: 50px 20px 20px;
            margin-top: 20px;
            width: 100%;
            position: relative;
            z-index: 1;
        }

        footer .container {
            max-width: 1200px;
            margin: 0 auto;
            background: transparent;
            box-shadow: none;
            border-radius: 0;
            overflow: visible;
        }

        .footer-content {
            display: grid;
            grid-template-columns: repeat(auto-fit, minmax(250px, 1fr));
            gap: 40px;
            margin-bottom: 40px;
        }

        .footer-section h3 {
            margin-bottom: 20px;
            color: #4ECDC4;
        }

        .footer-section ul {
            list-style: none;
        }

        .footer-section ul li {
            margin-bottom: 10px;
        }

        .footer-section a {
            color: #ecf0f1;
            text-decoration: none;
            transition: all 0.3s;
            display: inline-block;
        }

        .footer-section a:hover {
            color: #4ECDC4;
            transform: translateX(5px);
        }

        .footer-bottom {
            text-align: center;
            padding-top: 30px;
            border-top: 1px solid rgba(255,255,255,0.1);
            opacity: 0.8;
        }

        .social-links {
            display: flex;
            gap: 15px;
            margin-top: 15px;
        }

        .social-links a {
            width: 40px;
            height: 40px;
            background: rgba(255,255,255,0.1);
            border-radius: 50%;
            display: flex;
            align-items: center;
            justify-content: center;
            transition: all 0.3s;
        }

        .social-links a:hover {
            background: #4ECDC4;
            transform: translateY(-5px) rotate(360deg);
            box-shadow: 0 5px 15px rgba(78, 205, 196, 0.4);
        }

        /* Dataset Display Section */
        .dataset-section {
            margin-top: 40px;
        }

        .dataset-tabs {
            display: flex;
            gap: 10px;
            margin-bottom: 20px;
            border-bottom: 2px solid #ddd;
        }

        .dataset-tab {
            padding: 15px 30px;
            background: #f8f9fa;
            border: none;
            border-bottom: 3px solid transparent;
            cursor: pointer;
            font-weight: 600;
            color: #666;
            transition: all 0.3s;
        }

        .dataset-tab.active {
            color: #667eea;
            border-bottom-color: #667eea;
            background: white;
        }

        .dataset-tab:hover {
            color: #667eea;
        }

        .dataset-content {
            display: none;
        }

        .dataset-content.active {
            display: block;
        }

        .data-search {
            margin-bottom: 20px;
        }

        .data-search input {
            width: 100%;
            padding: 12px 15px;
            border: 2px solid #ddd;
            border-radius: 8px;
            font-size: 16px;
        }

        .data-search input:focus {
            outline: none;
            border-color: #667eea;
        }

        .data-samples {
            max-height: 600px;
            overflow-y: auto;
            padding-right: 10px;
        }

        .data-sample {
            background: #f8f9fa;
            padding: 20px;
            margin-bottom: 15px;
            border-radius: 8px;
            border-left: 4px solid #667eea;
            transition: all 0.3s;
        }

        .data-sample:hover {
            box-shadow: 0 5px 15px rgba(0,0,0,0.1);
            transform: translateX(5px);
        }

        .data-sample-header {
            display: flex;
            justify-content: space-between;
            align-items: center;
            margin-bottom: 10px;
        }

        .data-sample-id {
            font-weight: bold;
            color: #667eea;
            font-size: 1.1em;
        }

        .data-sample-entities {
            display: flex;
            flex-wrap: wrap;
            gap: 5px;
            margin-top: 10px;
        }

        .data-sample-text {
            line-height: 1.8;
            font-size: 16px;
            color: #333;
        }

        .data-stats {
            background: white;
            padding: 20px;
            border-radius: 8px;
            margin-bottom: 20px;
            display: grid;
            grid-template-columns: repeat(auto-fit, minmax(150px, 1fr));
            gap: 15px;
        }

        .data-stat-item {
            text-align: center;
            padding: 15px;
            background: #f8f9fa;
            border-radius: 8px;
        }

        .data-stat-item h4 {
            font-size: 2em;
            color: #667eea;
            margin-bottom: 5px;
        }

        .data-stat-item p {
            color: #666;
            font-size: 0.9em;
        }
    </style>
</head>
<body>
    <!-- Navigation Header -->
    <header>
        <div class="header-top">
            <div class="container">
                <div>
                    <a href="mailto:help@rskworld.in"><i class="fas fa-envelope"></i> help@rskworld.in</a>
                    <a href="tel:+919330539277"><i class="fas fa-phone"></i> +91 93305 39277</a>
                </div>
                <div>
                    <a href="https://rskworld.in" target="_blank"><i class="fas fa-globe"></i> rskworld.in</a>
                </div>
            </div>
        </div>
        <div class="header-main">
            <div class="container">
                <div class="logo">
                    <i class="fas fa-file-alt"></i>
                    <div class="logo-text">
                        <h1>NER Dataset</h1>
                        <p>Named Entity Recognition</p>
                    </div>
                </div>
                <nav>
                    <ul>
                        <li><a href="index.html">Home</a></li>
                        <li><a href="demo.html">Demo</a></li>
                        <li><a href="index.html#features">Features</a></li>
                        <li><a href="#dataset">Dataset</a></li>
                        <li><a href="index.html#technologies">Technologies</a></li>
                        <li><a href="README.md">Documentation</a></li>
                    </ul>
                </nav>
            </div>
        </div>
    </header>

    <div class="container">
        <div class="page-header">
            <h1><i class="fas fa-file-alt"></i> Named Entity Recognition Dataset</h1>
            <p>Interactive Demo - Extract entities from text</p>
        </div>

        <div class="content">
            <div class="section">
                <h2><i class="fas fa-keyboard"></i> Try It Out</h2>
                <div class="input-section">
                    <div class="input-group">
                        <label for="textInput">Enter text to analyze:</label>
                        <textarea id="textInput" placeholder="Enter text here... Example: Apple Inc. was founded by Steve Jobs in Cupertino, California on April 1, 1976.">Apple Inc. was founded by Steve Jobs in Cupertino, California on April 1, 1976.</textarea>
                    </div>
                    <button onclick="analyzeText()">
                        <i class="fas fa-search"></i> Analyze Entities
                    </button>
                </div>

                <div class="result-section">
                    <h3>Results:</h3>
                    <div id="highlightedText" class="highlighted-text">
                        Enter text above and click "Analyze Entities" to see results.
                    </div>

                    <div id="stats" class="stats" style="display: none;"></div>
                </div>

                <div class="legend">
                    <h3><i class="fas fa-tags"></i> Entity Types</h3>
                    <div class="legend-items">
                        <div class="legend-item">
                            <div class="legend-color entity-PERSON"></div>
                            <span>PERSON</span>
                        </div>
                        <div class="legend-item">
                            <div class="legend-color entity-ORG"></div>
                            <span>ORG</span>
                        </div>
                        <div class="legend-item">
                            <div class="legend-color entity-LOC"></div>
                            <span>LOC</span>
                        </div>
                        <div class="legend-item">
                            <div class="legend-color entity-DATE"></div>
                            <span>DATE</span>
                        </div>
                        <div class="legend-item">
                            <div class="legend-color entity-MONEY"></div>
                            <span>MONEY</span>
                        </div>
                        <div class="legend-item">
                            <div class="legend-color entity-PERCENT"></div>
                            <span>PERCENT</span>
                        </div>
                    </div>
                </div>
            </div>

            <div class="section">
                <h2><i class="fas fa-lightbulb"></i> Sample Texts</h2>
                <p>Click on any sample to analyze it:</p>
                <div class="sample-texts">
                    <div class="sample-card" onclick="loadSample('Apple Inc. was founded by Steve Jobs in Cupertino, California on April 1, 1976.')">
                        <strong>Tech Company</strong><br>
                        Apple Inc. was founded by Steve Jobs...
                    </div>
                    <div class="sample-card" onclick="loadSample('Elon Musk is the CEO of Tesla and SpaceX.')">
                        <strong>Business Leader</strong><br>
                        Elon Musk is the CEO of Tesla...
                    </div>
                    <div class="sample-card" onclick="loadSample('The conference will be held in New York on March 15, 2024.')">
                        <strong>Event</strong><br>
                        The conference will be held in New York...
                    </div>
                    <div class="sample-card" onclick="loadSample('Microsoft Corporation is headquartered in Redmond, Washington.')">
                        <strong>Corporation</strong><br>
                        Microsoft Corporation is headquartered...
                    </div>
                </div>
            </div>

            <div class="section">
                <h2><i class="fas fa-info-circle"></i> About This Dataset</h2>
                <p>This Named Entity Recognition dataset contains labeled entities including:</p>
                <ul style="margin-left: 20px; margin-top: 10px; line-height: 2;">
                    <li><strong>Persons:</strong> Names of people</li>
                    <li><strong>Organizations:</strong> Companies, institutions, and organizations</li>
                    <li><strong>Locations:</strong> Cities, countries, and geographical locations</li>
                    <li><strong>Dates:</strong> Time expressions and dates</li>
                    <li><strong>Money:</strong> Monetary values</li>
                    <li><strong>Percent:</strong> Percentage values</li>
                </ul>
                <p style="margin-top: 20px;">
                    The dataset is available in multiple formats (CSV, JSON, BIO) and is ready for training NER models using spaCy, NLTK, or Transformers.
                </p>
            </div>

            <div class="section dataset-section" id="dataset">
                <h2><i class="fas fa-database"></i> Complete Dataset</h2>
                <p>Browse all training and test samples with labeled entities:</p>
                
                <div class="dataset-tabs">
                    <button class="dataset-tab active" onclick="showDataset('train', this)">
                        <i class="fas fa-graduation-cap"></i> Training Data (50 samples)
                    </button>
                    <button class="dataset-tab" onclick="showDataset('test', this)">
                        <i class="fas fa-vial"></i> Test Data (30 samples)
                    </button>
                </div>

                <div id="trainData" class="dataset-content active">
                    <div class="data-search">
                        <input type="text" id="trainSearch" placeholder="Search training data..." onkeyup="filterData('train', this.value)">
                    </div>
                    <div id="trainStats" class="data-stats"></div>
                    <div id="trainSamples" class="data-samples"></div>
                </div>

                <div id="testData" class="dataset-content">
                    <div class="data-search">
                        <input type="text" id="testSearch" placeholder="Search test data..." onkeyup="filterData('test', this.value)">
                    </div>
                    <div id="testStats" class="data-stats"></div>
                    <div id="testSamples" class="data-samples"></div>
                </div>
            </div>
        </div>

    </div>

    <!-- Footer -->
    <footer class="footer">
        <div class="container">
            <div class="footer-content">
                <div class="footer-section">
                    <h3>About</h3>
                    <p>Named Entity Recognition Dataset for information extraction and NLP model training. Created by RSK World.</p>
                    <div class="social-links">
                        <a href="https://rskworld.in" target="_blank" title="Website"><i class="fas fa-globe"></i></a>
                        <a href="mailto:help@rskworld.in" title="Email"><i class="fas fa-envelope"></i></a>
                        <a href="tel:+919330539277" title="Phone"><i class="fas fa-phone"></i></a>
                    </div>
                </div>
                <div class="footer-section">
                    <h3>Quick Links</h3>
                    <ul>
                        <li><a href="index.html">Home</a></li>
                        <li><a href="demo.html">Demo</a></li>
                        <li><a href="README.md">Documentation</a></li>
                        <li><a href="index.html#features">Features</a></li>
                    </ul>
                </div>
                <div class="footer-section">
                    <h3>Resources</h3>
                    <ul>
                        <li><a href="dataset/train.json">Training Data</a></li>
                        <li><a href="dataset/test.json">Test Data</a></li>
                        <li><a href="scripts/">Scripts</a></li>
                        <li><a href="requirements.txt">Requirements</a></li>
                    </ul>
                </div>
                <div class="footer-section">
                    <h3>Contact</h3>
                    <ul>
                        <li><i class="fas fa-globe"></i> <a href="https://rskworld.in" target="_blank">rskworld.in</a></li>
                        <li><i class="fas fa-envelope"></i> <a href="mailto:help@rskworld.in">help@rskworld.in</a></li>
                        <li><i class="fas fa-phone"></i> <a href="tel:+919330539277">+91 93305 39277</a></li>
                    </ul>
                </div>
            </div>
            <div class="footer-bottom">
                <p>&copy; 2026 <strong>RSK World</strong> - Free Programming Resources & Source Code. All rights reserved.</p>
                <p style="margin-top: 10px; font-size: 0.9em;">
                    Licensed under <a href="LICENSE" style="color: #4ECDC4;">MIT License</a>
                </p>
            </div>
        </div>
    </footer>

    <script>
        // Simple NER patterns (for demo purposes)
        const patterns = {
            PERSON: [
                /\b([A-Z][a-z]+ [A-Z][a-z]+)\b/g,  // First Last
                /\b([A-Z][a-z]+ [A-Z]\. [A-Z][a-z]+)\b/g  // First M. Last
            ],
            ORG: [
                /\b([A-Z][a-z]+ (?:Inc\.|Corporation|Corp\.|LLC|Ltd\.|Company|Co\.))\b/g,
                /\b([A-Z][A-Z]+)\b/g,  // Acronyms like NASA, IBM
                /\b(Apple|Microsoft|Google|Amazon|Facebook|Twitter|Tesla|SpaceX|Netflix|Uber|Airbnb|Spotify|LinkedIn|Zoom|Dropbox|Salesforce|Oracle|Intel|Adobe|NVIDIA|OpenAI|Meta|Alphabet)\b/g
            ],
            LOC: [
                /\b([A-Z][a-z]+(?:, [A-Z][a-z]+)?)\b/g,  // City, State
                /\b(New York|Los Angeles|San Francisco|Chicago|Houston|Phoenix|Philadelphia|San Antonio|San Diego|Dallas|San Jose|Austin|Jacksonville|Fort Worth|Columbus|Charlotte|San Francisco|Indianapolis|Seattle|Denver|Washington|Boston|El Paso|Detroit|Nashville|Portland|Oklahoma City|Las Vegas|Memphis|Louisville|Baltimore|Milwaukee|Albuquerque|Tucson|Fresno|Sacramento|Kansas City|Mesa|Atlanta|Omaha|Colorado Springs|Raleigh|Virginia Beach|Miami|Oakland|Minneapolis|Tulsa|Cleveland|Wichita|Arlington)\b/g,
                /\b(United States|USA|UK|Canada|Mexico|China|Japan|India|Germany|France|Italy|Spain|Brazil|Australia|Russia|South Korea|Netherlands|Sweden|Switzerland|Belgium|Norway|Denmark|Finland|Poland|Austria|Greece|Portugal|Ireland|Czech Republic|Romania|Hungary)\b/g
            ],
            DATE: [
                /\b(January|February|March|April|May|June|July|August|September|October|November|December) \d{1,2}, \d{4}\b/g,
                /\b\d{4}\b/g,  // Years
                /\b(January|February|March|April|May|June|July|August|September|October|November|December) \d{4}\b/g
            ],
            MONEY: [
                /\$\d+(?:,\d{3})*(?:\.\d{2})?\b/g,
                /\b\d+(?:,\d{3})* (?:dollars|USD)\b/gi
            ],
            PERCENT: [
                /\b\d+(?:\.\d+)?%\b/g
            ]
        };

        function analyzeText() {
            const text = document.getElementById('textInput').value;
            if (!text.trim()) {
                alert('Please enter some text to analyze.');
                return;
            }

            const entities = extractEntities(text);
            displayResults(text, entities);
        }

        function extractEntities(text) {
            const entities = [];
            const entityMap = new Map();

            // Extract entities for each type
            for (const [type, patternList] of Object.entries(patterns)) {
                for (const pattern of patternList) {
                    let match;
                    while ((match = pattern.exec(text)) !== null) {
                        const start = match.index;
                        const end = start + match[0].length;
                        const entityText = match[0];

                        // Check if this position is already covered
                        let overlap = false;
                        for (const [existingStart, existingEnd] of entityMap.keys()) {
                            if (!(end <= existingStart || start >= existingEnd)) {
                                overlap = true;
                                break;
                            }
                        }

                        if (!overlap) {
                            entities.push({
                                text: entityText,
                                type: type,
                                start: start,
                                end: end
                            });
                            entityMap.set([start, end], type);
                        }
                    }
                }
            }

            // Sort by start position
            entities.sort((a, b) => a.start - b.start);
            return entities;
        }

        function displayResults(text, entities) {
            let highlighted = text;
            const stats = {};

            // Highlight entities (process in reverse to maintain indices)
            for (let i = entities.length - 1; i >= 0; i--) {
                const entity = entities[i];
                const before = highlighted.substring(0, entity.start);
                const after = highlighted.substring(entity.end);
                const highlightedEntity = `<span class="entity entity-${entity.type}" title="${entity.type}">${entity.text}</span>`;
                highlighted = before + highlightedEntity + after;

                // Count entities
                stats[entity.type] = (stats[entity.type] || 0) + 1;
            }

            document.getElementById('highlightedText').innerHTML = highlighted || '<em>No entities found.</em>';

            // Display statistics
            const statsDiv = document.getElementById('stats');
            if (Object.keys(stats).length > 0) {
                statsDiv.style.display = 'grid';
                statsDiv.innerHTML = Object.entries(stats).map(([type, count]) => `
                    <div class="stat-card">
                        <h3>${count}</h3>
                        <p>${type}</p>
                    </div>
                `).join('');
            } else {
                statsDiv.style.display = 'none';
            }
        }

        function loadSample(text) {
            document.getElementById('textInput').value = text;
            analyzeText();
        }

        // Analyze on page load
        window.onload = function() {
            analyzeText();
            loadDataset();
        };

        // Dataset loading and display
        let trainData = [];
        let testData = [];

        function highlightEntities(text, entities) {
            if (!entities || entities.length === 0) return text;
            
            // Sort entities by start position (reverse to avoid index shifting)
            const sortedEntities = [...entities].sort((a, b) => b.start - a.start);
            let highlighted = text;
            
            for (const entity of sortedEntities) {
                const before = highlighted.substring(0, entity.start);
                const after = highlighted.substring(entity.end);
                const entityText = highlighted.substring(entity.start, entity.end);
                const highlightedEntity = `<span class="entity entity-${entity.label}" title="${entity.label}">${entityText}</span>`;
                highlighted = before + highlightedEntity + after;
            }
            
            return highlighted;
        }

        function loadDataset() {
            // Load training data
            fetch('dataset/train.json')
                .then(response => response.json())
                .then(data => {
                    trainData = data;
                    displayDataset('train', data);
                    updateStats('train', data);
                })
                .catch(error => {
                    console.error('Error loading training data:', error);
                    document.getElementById('trainSamples').innerHTML = '<p style="color: red;">Error loading training data. Please ensure dataset/train.json exists.</p>';
                });

            // Load test data
            fetch('dataset/test.json')
                .then(response => response.json())
                .then(data => {
                    testData = data;
                    displayDataset('test', data);
                    updateStats('test', data);
                })
                .catch(error => {
                    console.error('Error loading test data:', error);
                    document.getElementById('testSamples').innerHTML = '<p style="color: red;">Error loading test data. Please ensure dataset/test.json exists.</p>';
                });
        }

        function displayDataset(type, data) {
            const container = document.getElementById(type + 'Samples');
            if (!container) return;

            let html = '';
            data.forEach((sample, index) => {
                const highlighted = highlightEntities(sample.text, sample.entities);
                const entityList = sample.entities.map(e => 
                    `<span class="entity entity-${e.label}" title="${e.label}">${e.text}</span>`
                ).join(' ');

                html += `
                    <div class="data-sample" data-index="${index}">
                        <div class="data-sample-header">
                            <span class="data-sample-id">Sample #${sample.id || index + 1}</span>
                            <span style="color: #666; font-size: 0.9em;">${sample.entities.length} entities</span>
                        </div>
                        <div class="data-sample-text">${highlighted}</div>
                        <div class="data-sample-entities">
                            <strong>Entities:</strong> ${entityList || 'None'}
                        </div>
                    </div>
                `;
            });

            container.innerHTML = html;
        }

        function updateStats(type, data) {
            const statsContainer = document.getElementById(type + 'Stats');
            if (!statsContainer) return;

            const totalSamples = data.length;
            const totalEntities = data.reduce((sum, sample) => sum + (sample.entities?.length || 0), 0);
            const entityTypes = {};
            
            data.forEach(sample => {
                sample.entities?.forEach(entity => {
                    entityTypes[entity.label] = (entityTypes[entity.label] || 0) + 1;
                });
            });

            const uniqueTypes = Object.keys(entityTypes).length;

            statsContainer.innerHTML = `
                <div class="data-stat-item">
                    <h4>${totalSamples}</h4>
                    <p>Total Samples</p>
                </div>
                <div class="data-stat-item">
                    <h4>${totalEntities}</h4>
                    <p>Total Entities</p>
                </div>
                <div class="data-stat-item">
                    <h4>${uniqueTypes}</h4>
                    <p>Entity Types</p>
                </div>
                <div class="data-stat-item">
                    <h4>${(totalEntities / totalSamples).toFixed(1)}</h4>
                    <p>Avg per Sample</p>
                </div>
            `;
        }

        function showDataset(type, element) {
            // Update tabs
            document.querySelectorAll('.dataset-tab').forEach(tab => tab.classList.remove('active'));
            if (element) {
                element.classList.add('active');
            } else {
                // Fallback: find the clicked button
                const buttons = document.querySelectorAll('.dataset-tab');
                buttons.forEach(btn => {
                    if (btn.textContent.includes(type === 'train' ? 'Training' : 'Test')) {
                        btn.classList.add('active');
                    }
                });
            }

            // Update content
            document.querySelectorAll('.dataset-content').forEach(content => content.classList.remove('active'));
            document.getElementById(type + 'Data').classList.add('active');
        }

        function filterData(type, searchTerm) {
            const data = type === 'train' ? trainData : testData;
            if (!data) return;

            const filtered = data.filter(sample => {
                const text = sample.text.toLowerCase();
                const search = searchTerm.toLowerCase();
                return text.includes(search) || 
                       sample.entities.some(e => e.text.toLowerCase().includes(search) || e.label.toLowerCase().includes(search));
            });

            displayDataset(type, filtered);
        }
    </script>
</body>
</html>

1,120 lines•38.2 KB

markup

Theme Settings

Color Scheme

Display Options

Font Size