Petuum
Petuum is a distributed machine learning framework. It aims to provide a generic algorithmic and systems interface to large scale machine learning, and takes care of difficult systems "plumbing work" and algorithmic acceleration, while simplifying the distributed implementation of ML programs - allowing you to focus on model perfection and Big Data Analytics. Petuum runs efficiently at scale on re
Hive/Pigを使ã£ãŸKDD'12 track2ã®åºƒå‘Šã‚¯ãƒªãƒƒã‚¯çŽ‡äºˆæ¸¬
1. Hive/Pigを使ã£ãŸKDD'12 track2 ã®åºƒå‘Šã‚¯ãƒªãƒƒã‚¯çŽ‡äºˆæ¸¬ 油井 èª m.yui@aist.go.jp 産æ¥æŠ€è¡“ç·åˆç ”究所 æƒ…å ±æŠ€è¡“ç ”ç©¶éƒ¨é–€ Twitter ID: @myui スライド http://www.slideshare.net/myui/dsirnlp-myuilt 1 http://goo.gl/Ulf3A 2. KDDcup 2012 track2 • 検索ãƒã‚°ã‚’基ã«ã€æ¤œç´¢ã‚¨ãƒ³ã‚¸ãƒ³ã®åºƒå‘Šã®ã‚¯ãƒªãƒƒã‚¯ 率(Click-Through Rate)を推定ã™ã‚‹ã‚¿ã‚¹ã‚¯ – ä¸å›½ã®3大検索エンジンã®ä¸€ã¤soso.comã®å®Ÿãƒ‡ãƒ¼ã‚¿ • 検索語ãªã©ã¯Hash値ãªã©ã‚’利用ã—ã¦ã™ã¹ã¦æ•°å€¤åŒ–ã•ã‚Œã¦ã„ ã‚‹ – Trainingデータ(ç´„10GB+2.2GB, 15億レコード） – Testデータ（約1.3GB, 2億レコード） • å¦ç¿’データã®1.33割ãŒè©•ä¾¡ç”¨ãƒ‡ãƒ¼ã‚¿ã‚»ãƒƒãƒˆ –
Cloudera: 機械å¦ç¿’ãŒåºƒãŒã‚‹ä¸ã§ã®ãƒ‡ãƒ¼ã‚¿ãƒ¢ãƒ‡ãƒ« - ワザノム| wazanova
Josh Willsã¯ã€Googleã§åºƒå‘Šã‚ªãƒ¼ã‚¯ã‚·ãƒ§ãƒ³ã‚·ã‚¹ãƒ†ãƒ ã¨ãƒ‡ãƒ¼ã‚¿åˆ†æžã‚¤ãƒ³ãƒ•ãƒ©ã®é–‹ç™ºã‚’担当ã—ã€ç¾åœ¨ã¯ã€Clouderaã®Senior Director of Data Scienceを勤ã‚ã¦ã„ã¾ã™ã€‚ 機械å¦ç¿’ã«ã‚ˆã‚‹ãƒ‡ãƒ¼ã‚¿åˆ†æžãŒã€Googleãªã©ã«é™ã‚‰ãšã€Airbnbã‚„Etsyãªã©ãã®ä»–ã®ãƒãƒƒãƒˆä¼æ¥ã«ã‚‚広ãŒã£ã¦ã„ãä¸ã€ä»Šå¾Œå¿…è¦ã¨ã•ã‚Œã‚‹ãƒ‡ãƒ¼ã‚¿ãƒ¢ãƒ‡ãƒ«ã«ã¤ã„㦠ã€Midwest.io 㨠GraphLab Conference ã§è¬›æ¼”ã—ã¦ã„ã¾ã™ã€‚ 複数ã®æ©Ÿæ¢°å¦ç¿’分æžã‚’æ“ã‚‹Googleã®ãƒãƒ£ãƒ¬ãƒ³ã‚¸ 機械å¦ç¿’分æžã®ãƒ™ãƒ¼ã‚¹ã¨ãªã‚‹å„サービスã®ä»•æ§˜ãŒé »ç¹ã«å¤‰æ›´ã«ãªã‚‹ã€‚ 機械å¦ç¿’分æžã¯ãã‚Œãžã‚Œå˜ä½“ã§ã‚‚複雑ãªã®ã«ã€Googleã®å ´åˆã€ãれらãŒç›¸äº’ã«é–¢é€£ãŒã‚るケースãŒå¤šã‹ã£ãŸã€‚一ã¤ã®æ©Ÿæ¢°å¦ç¿’分æžã®å¤‰æ›´ãŒã€ä»–ã®æ©Ÿæ¢°å¦ç¿’ã«ã©ã†å½±éŸ¿ãŒã§ã‚‹ã‹ãŒæŠŠæ¡ã—ãã‚Œãªã„。ãã®çµæžœã€åºƒå‘Šé…信システムをç†è§£ã§ããªã„（= コントãƒãƒ¼ãƒ«
都立大 自然言語処ç†ç ”究室 - 自然言語処ç†ã‚’å¦ã¶æŽ¨è–¦æ›¸ç±
自然言語処ç†ã‚’å¦ã¶æŽ¨è–¦æ›¸ç±ã‚’紹介ã—ã¾ã™ã€‚2021å¹´03月ç¾åœ¨ã€è‡ªç„¶è¨€èªžå‡¦ç†ã‚’勉強ã—ãŸã„ç†å·¥ç³»ã®å¦ç”Ÿãƒ»ã‚¨ãƒ³ã‚¸ãƒ‹ã‚¢ã®äººã¯ã€ä»¥ä¸‹ã®æœ¬ã‚’推薦ã—ã¾ã™ã€‚ （概è¦ï¼‰è‡ªç„¶è¨€èªžå‡¦ç†ï¼ˆæ”¾é€å¤§å¦å‡ºç‰ˆï¼‰ （ç†è«–）言語処ç†ã®ãŸã‚ã®æ©Ÿæ¢°å¦ç¿’入門＋深層å¦ç¿’ã«ã‚ˆã‚‹è‡ªç„¶è¨€èªžå‡¦ç† （実装）Python 機械å¦ç¿’プãƒã‚°ãƒ©ãƒŸãƒ³ã‚° 第3版 自然言語処ç†ã‚’勉強ã—ãŸã„ã€éžç†å·¥ç³»ãƒ»éžã‚¨ãƒ³ã‚¸ãƒ‹ã‚¢ã®äººã«ã¯ã€ä»¥ä¸‹ã®æœ¬ã‚’推薦ã—ã¾ã™ã€‚ （数å¼ãªã—）自然言語処ç†ã®åŸºæœ¬ã¨æŠ€è¡“ （数å¼ã‚り）自然言語処ç†ï¼ˆæ”¾é€å¤§å¦å‡ºç‰ˆï¼‰ オライリーã‹ã‚‰å‡ºã¦ã„る「入門 自然言語処ç†ã€ã¯ç‰¹æ®Šãªæœ¬ï¼ˆè©³ã—ã„人ãŒã“れを使ã£ã¦ãƒ¬ã‚¯ãƒãƒ£ãƒ¼ã—ã¦ãれるãªã‚‰ã¨ã‚‚ã‹ãã€ç‹¬ç¿’ã«å‘ã„ã¦ã„ãªã„）ã§ã€Python 2 ã§æ›¸ã‹ã‚Œã¦ã„ã‚‹ã ã‘ã§ãªãã€ã™ã§ã«å‹•ã‹ãªããªã£ãŸã‚³ãƒ¼ãƒ‰ã‚‚多々ã‚ã‚Šã€2019年時点ã§ã¯èªã¾ãªã„æ–¹ãŒã„ã„本ã§ã™ã€‚（ãã‚Œã§ã‚‚ã©ã†ã—ã¦ã‚‚ã€æ„地ã§ã‚‚èªã¿ãŸã„人ã¯ã€æœ¬å®¶ã«ã‚ã‚‹ Python 3 対応
Machine Learning that Matters(ICML 2012) èªã‚“ã - 糞糞糞ãƒãƒƒãƒˆå¼æ…¶
機械å¦ç¿’ã‚„ã£ã¦ã‚‹äººã¯çš†èªã‚€ã¹ãã ã¨æ€ã†ï¼Ž Machine Learning that Matters (pdf) æ¦‚è¦ æ©Ÿæ¢°å¦ç¿’ã®ãƒˆãƒƒãƒ—カンファレンスICMLã«æ•°å¼/アルゴリズム/定ç†ã‚’1ã¤ã‚‚書ã‹ãšã«é€šã£ãŸè«–文． 機械å¦ç¿’ã¯ä½•ã®ãŸã‚ã«é‡è¦ãªã®ã‹ï¼Œç¾å®Ÿä¸–ç•Œã¨ã®ç¹‹ãŒã‚Šã‚’失ã£ã¦ã„ãªã„ã‹ï¼Œã‚ãªãŸã¯ã€Œæ©Ÿæ¢°å¦ç¿’ã£ã¦ä½•ã®å½¹ã«ç«‹ã¤ã®?ã€ã¨èžã‹ã‚ŒãŸæ™‚ã«ã¡ã‚ƒã‚“ã¨ç”ãˆã‚‰ã‚Œã‚‹ã®ã‹ï¼Œã¾ã å–り組ã¾ã‚Œã¦ãªã„é‡è¦ãªå•é¡Œã¯ä½•ã‹ï¼Œã«ã¤ã„ã¦è§¦ã‚ŒãŸé•·æ–‡ãƒ–ãƒã‚°ã®ã‚ˆã†ãªè«–文． contributions㯠機械å¦ç¿’ã®ç ”究ã¨äººé¡žã¨ç§‘å¦å…¨ä½“ã«ãŠã‘ã‚‹ã‚‚ã£ã¨å¤§ããªç ”究ã¨ã®é–“ã«ã‚ã‚‹æ¬ è½ã«å¯¾ã™ã‚‹æ˜Žç¢ºãªç‰¹å®šã¨è§£èª¬ ãã®ã‚®ãƒ£ãƒƒãƒ—ã«å–り組むãŸã‚ã®ç¬¬ä¸€æ© (ã©ã†è¨³ã—ã¦ã„ã„ã‹ã‚ã‹ã‚‰ãªã‹ã£ãŸ) 機械å¦ç¿’ã«ãŠã„ã¦é‡è¦ãªå•é¡Œã®æ示 機械å¦ç¿’ã®ãŸã‚ã®æ©Ÿæ¢°å¦ç¿’(è¦ç´„: マッシュルームやアヤメã®åˆ†é¡žå™¨ã‚’æ»ã¬ã»ã©ä½œã£ã¦ä½•ã®å½¹ã«ç«‹ã£ãŸã®ï¼Ÿ) ベンãƒãƒžãƒ¼ã‚¯ãƒ‡ãƒ¼ã‚¿ã®å•é¡Œ ã“
Introducing SAMOA, an open source platform for mining big data streams.
https://github.com/yahoo/samoa Machine learning and data mining are well established techniques in the world of IT and especially among web companies and startups. Spam detection, personalization and recommendations are just a few of the applications made possible by mining the huge quantity of data available nowadays. However, “big data†is not only about Volume, but also about Velocity (and Vari
A use case of online machine learning using Jubatus
This document summarizes a presentation about using online machine learning with Jubatus for a real estate search service called SUUMO. Toru Shimogaki from NTT DATA Corporation presented on how SUUMO uses Jubatus classifier and multidimensional scaling to learn users' preferences and provide personalized search results. The system presents users with two candidate properties and learns their taste
Vowpal Wabbit (Fast Learning)
This is a project started at Yahoo! Research and continuing at Microsoft Research to design a fast, scalable, useful learning algorithm. VW is the essence of speed in machine learning, able to learn from terafeature datasets with ease. Via parallel learning, it can exceed the throughput of any single machine network interface when doing linear learning, a first amongst learning algorithms. We prim
10分ã§åˆ†ã‹ã‚‹Jubatus/ML/NLP/ãã®ä»–
10分ã§ã‚ã‹ã‚‹Jubatus/ML/NLP/ãã®ä»–
1
ランã‚ング
ランã‚ング
ランã‚ング
リリースã€éšœå®³æƒ…å ±ãªã©ã®ã‚µãƒ¼ãƒ“スã®ãŠçŸ¥ã‚‰ã›
最新ã®äººæ°—エントリーã®é…ä¿¡
j次ã®ãƒ–ックマーク
kå‰ã®ãƒ–ックマーク
lã‚ã¨ã§èªã‚€
eコメント一覧を開ã
oページを開ã
Kaggle: Your Machine Learning and Data Science Community
Elements of Statistical Learning: data mining, inference, and prediction. 2nd Edition.
{{#tags}}- {{label}}
{{/tags}}