MinHash and Locality-Sensitive Hashing (LSH) near-duplicate document clustering and Jaccard similarity estimation.
genpark-minhash-lsh-document-deduplicator-skill
MinHash and Locality-Sensitive Hashing (LSH) near-duplicate document clustering and Jaccard similarity estimation.
Part of the GenPark AI Agent Skills Matrix. Production-ready, zero external dependencies, native Python 3.9+ standard library.
Features - Zero Third-Party Dependencies: Pure Python standard library (collections, re, math). - High-Performance NLP: Subword BPE merges, Okapi BM25 ranking, and Damerau-Levenshtein metrics. - Native MCP Protocol Support: Integrated JSON-RPC 2.0 stdio server ready for Claude Desktop, Cursor, and Windsurf.
Add to your claudedesktopconfig.json or cursor.json:
License MIT License. Copyright (c) 2026 AlphaPark Inc. & Alpha-Park.
From the project README.
Add the radar badge to your README — it shows your project was picked up by MCP Radar and links to this page:
[](https://mcp.liqiwa.com/s/Alpha-Park--genpark-minhash-lsh-document-deduplicator-skill.html)
MinHash and Locality-Sensitive Hashing (LSH) near-duplicate document clustering and Jaccard similarity estimation.
alphaparkinc/genpark-bpe-tokenizer-byte-pair-encoder-skillByte-Pair Encoding (BPE) subword tokenizer trainer and deterministic text encoder/decoder.
Alpha-Park/genpark-bpe-tokenizer-byte-pair-encoder-skillByte-Pair Encoding (BPE) subword tokenizer trainer and deterministic text encoder/decoder.
Alpha-Park/genpark-levenshtein-damerau-fuzzy-matcher-skillDamerau-Levenshtein edit distance metric with transposition detection and fuzzy string candidate ranker.
korovin-aa97/talkthrough-mcpMCP server: turn narrated screen recordings into agent-ready data — local Whisper transcript, scene keyframes, OCR, wall-clock anchoring. Record your screen, talk — your AI agent files the bugs.
alphaparkinc/genpark-role-playing-autonomous-agent-crew-orchestrator-skillRole-playing autonomous agent orchestrator managing crews (CrewAI style)
The top new MCP servers of the week, every Monday. No spam, unsubscribe anytime.