Spaces:
Running
Running
| <html lang="en"><head><meta charset="utf-8"><meta name="viewport" content="width=device-width,initial-scale=1"><meta name="description" content="ResearchLens multilingual document intelligence"><title>ResearchLens · Multilingual Document Intelligence</title> | |
| <style> | |
| :root{font-family:Inter,ui-sans-serif,system-ui,-apple-system,BlinkMacSystemFont,"Segoe UI",sans-serif;background:#08111c;color:#eef6ff;line-height:1.5}*{box-sizing:border-box}body{margin:0;background:radial-gradient(circle at 85% 0,#153662 0,transparent 31%),#08111c}.wrap{width:min(1160px,calc(100% - 32px));margin:auto}.hero{padding:64px 0 30px}.eyebrow{letter-spacing:.16em;font-size:.75rem;color:#79c5ff;font-weight:800}h1{font-size:clamp(2.5rem,6vw,5rem);line-height:1;margin:.5rem 0}.lead{max-width:800px;color:#abc0d5;font-size:1.08rem}.grid{display:grid;grid-template-columns:1.15fr .85fr;gap:18px}.panel{background:#0d1928;border:1px solid #203650;border-radius:18px;padding:20px}.panel h2{margin-top:0}.textarea{width:100%;min-height:280px;background:#07101b;color:#eef6ff;border:1px solid #2a405b;border-radius:12px;padding:14px;font:inherit;resize:vertical}.file{width:100%;margin:12px 0}.actions,.examples{display:flex;gap:9px;flex-wrap:wrap}.btn{border:0;border-radius:10px;padding:10px 14px;font-weight:750;cursor:pointer}.primary{background:#2d6bff;color:#fff}.secondary{background:#18273a;color:#cfe3f7}.status{color:#90a9c0;font-size:.9rem;margin-top:10px}.results{padding:22px 0 60px}.metrics{display:grid;grid-template-columns:repeat(4,1fr);gap:10px}.metric{background:#0d1928;border:1px solid #203650;border-radius:14px;padding:14px}.metric span{display:block;color:#8fa8bf;font-size:.78rem}.metric strong{font-size:1.25rem}.resultgrid{display:grid;grid-template-columns:1fr 1fr;gap:18px;margin-top:18px}.summary{white-space:pre-wrap;color:#d8e7f4}.barrow{display:grid;grid-template-columns:150px 1fr 45px;gap:10px;align-items:center;margin:9px 0;font-size:.88rem}.bar{height:9px;background:#17283a;border-radius:999px;overflow:hidden}.bar i{display:block;height:100%;background:linear-gradient(90deg,#2d6bff,#71d0ff)}.muted{color:#8fa8bf}.method{margin:18px 0 55px;color:#9eb4c8}.badge{display:inline-block;padding:6px 9px;border:1px solid #2b4765;border-radius:999px;color:#a9d5f5;font-size:.78rem;margin:4px 5px 0 0}@media(max-width:820px){.grid,.resultgrid{grid-template-columns:1fr}.metrics{grid-template-columns:1fr 1fr}.barrow{grid-template-columns:110px 1fr 35px}} | |
| </style> | |
| <script src="https://cdnjs.cloudflare.com/ajax/libs/pdf.js/3.11.174/pdf.min.js"></script></head><body><header class="wrap hero"><div class="eyebrow">RODRIGO PANTOJA · APPLIED NLP</div><h1>🔎 ResearchLens</h1><p class="lead">Multilingual document intelligence for academic, policy and business writing. Analyze English, Spanish or Portuguese text directly in your browser, with no generative API.</p><span class="badge">Browser NLP</span><span class="badge">No server compute</span><span class="badge">Privacy-first</span><span class="badge">Static Hugging Face Space</span></header> | |
| <main class="wrap"><section class="grid"><div class="panel"><h2>Analyze a document</h2><textarea id="text" class="textarea" placeholder="Paste research, policy, business or academic text here..."></textarea><input id="file" class="file" type="file" accept=".pdf,.txt,.md"><div class="actions"><button id="analyze" class="btn primary">Analyze document</button><button id="clear" class="btn secondary">Clear</button></div><div id="status" class="status">Ready.</div></div><div class="panel"><h2>What it returns</h2><p class="muted">Document metrics, extractive summary, lexical signals and topic indicators. The methodology is intentionally transparent and lightweight.</p><h3>Try an example</h3><div class="examples"><button class="btn secondary" data-example="es">Spanish</button><button class="btn secondary" data-example="en">English</button><button class="btn secondary" data-example="pt">Portuguese</button></div><h3>Privacy</h3><p class="muted">Text analysis runs locally in your browser. No external generative API is called. PDF text extraction uses PDF.js in the browser.</p></div></section> | |
| <section id="results" class="results" hidden><div id="metrics" class="metrics"></div><div class="resultgrid"><div class="panel"><h2>Extractive summary</h2><div id="summary" class="summary"></div></div><div class="panel"><h2>Topic signals</h2><div id="topics"></div></div><div class="panel"><h2>Top lexical signals</h2><div id="keywords"></div></div><div class="panel"><h2>Methodology</h2><p class="muted">Language detection uses transparent marker scoring. Keywords combine unigram and bigram frequencies after stopword filtering. The summary scores original sentences against document-level lexical signals. Topic scores use a curated multilingual domain lexicon.</p></div></div><p class="method"><strong>Limitations:</strong> ResearchLens is an analytical aid, not a factual verifier. Scanned PDFs require OCR, mixed-language texts can confuse language detection, and heuristic summaries may miss nuance in highly technical documents.</p></section></main> | |
| <script src="engine.js"></script><script>if(window.pdfjsLib)pdfjsLib.GlobalWorkerOptions.workerSrc='https://cdnjs.cloudflare.com/ajax/libs/pdf.js/3.11.174/pdf.worker.min.js';</script><script src="app.js"></script></body></html> | |