[B! ãƒ™ã‚¤ã‚¸ã‚¢ãƒ³] katotakuã®ãƒ

katotaku id:katotaku

ãƒ™ã‚¤ã‚¸ã‚¢ãƒ³ã«é–¢ã™ã‚‹katotakuã®ãƒ–ãƒƒã‚¯ãƒžãƒ¼ã‚¯ (23)

${{author_name}}$

{{{comment_expanded}}}

{{label}}

{{#is_bookmark}}ãƒªã‚¹ãƒˆ{{/is_bookmark}}{{^is_bookmark}}ãƒªãƒ³ã‚¯{{/is_bookmark}}

${{author_name}}$
{{author_name}}{{created}}
{{ #comment }}{{ comment }}{{ /comment }}
- {{ label }}

${{author_name}}$

{{{comment_expanded}}}

{{label}}

{{#is_bookmark}}ãƒªã‚¹ãƒˆ{{/is_bookmark}}{{^is_bookmark}}ãƒªãƒ³ã‚¯{{/is_bookmark}}

Bayesian Setsã«ã‚ˆã‚‹é–¢é€£æ–‡æ›¸æ¤œç´¢ã‚·ã‚¹ãƒ†ãƒ Stupa - mixi engineer blog
éƒ½ä¼šã‚ˆã‚Šã‚‚ç”°èˆŽãŒå¥½ããªfujisawaã§ã™ã€‚Bayesian Setsã¨ã„ã†ã‚¢ãƒ«ã‚´ãƒªã‚ºãƒ ã‚’ä½¿ã£ã¦ã€é–¢é€£ã™ã‚‹æ–‡æ›¸ã‚’é«˜é€Ÿãƒ»é«˜ç²¾åº¦ã«æ¤œç´¢ã§ãã‚‹ã‚·ã‚¹ãƒ†ãƒ ã‚’ä½œæˆã—ã¾ã—ãŸã®ã§ã€ãã®ã”ç´¹ä»‹ã‚’ã•ã›ã¦ã„ãŸã ãã¾ã™ã€‚ Bayesian Setsã¨ã¯ Bayesian Setsã¯ã„ãã¤ã‹ã‚¢ã‚¤ãƒ†ãƒ ã‚’å…¥åŠ›ã™ã‚‹ã¨ã€ãã‚Œã‚’è£œå®Œã™ã‚‹ã‚ˆã†ãªã‚¢ã‚¤ãƒ†ãƒ ã‚’è¿”ã—ã¦ãã‚Œã‚‹ã‚¢ãƒ«ã‚´ãƒªã‚ºãƒ ã§ã™ã€‚åŽŸè‘—è«–æ–‡ã®å…ˆé ã«"Inspired by Google Sets"ã¨æ›¸ã‹ã‚Œã¦ã„ã‚‹ã‚ˆã†ã«ã€Google Setsã‚’å‚è€ƒã«ã—ã¦ä½œæˆã•ã‚Œã¦ã„ã¾ã™ã€‚å®Ÿéš›ã«ã©ã®ã‚ˆã†ãªå‡ºåŠ›ãŒå¾—ã‚‰ã‚Œã‚‹ã‹ã€Google Setsã«ä»¥ä¸‹ã®è¡¨ã®ã‚¯ã‚¨ãƒªã‚’æ¤œç´¢ã—ã¦è©¦ã—ã¦ã¿ã¾ã™ã¨ã€ ã‚¯ã‚¨ãƒª å‡ºåŠ› apple, banana chocolate, strawberry, vanilla, cherry, ... apple, macintosh software, windows, mac,
katotaku 2010/02/15
è‡ªç„¶è¨€èªžå‡¦ç†

algorithm

ãƒ™ã‚¤ã‚¸ã‚¢ãƒ³
ãƒªãƒ³ã‚¯
ãƒ™ã‚¤ã‚¸ã‚¢ãƒ³ãƒ•ã‚£ãƒ«ã‚¿ãƒ¼ã€€Perlã§ä½œã‚ŠãŸã„äººã«æ•™ãˆã¦ã‚ã’ãŸã„ã¡ã‚‡ã£ã¨ã—ãŸã“ã¨ - ãƒ—ãƒã‚°ãƒ©ãƒžã§ã‚ã‚ŠãŸã„
æ˜¨æ—¥ã®ã¯ã¦ãªã®ãƒ›ãƒƒãƒˆã‚¨ãƒ³ãƒˆãƒªãƒ¼ã«ã€Žå…¥é–€ãƒ™ã‚¤ã‚ºçµ±è¨ˆã€ã®èªã¿ã©ã“ã‚ã¨ã„ã†è¨˜äº‹ãŒè¼‰ã£ã¦ã„ã¾ã—ãŸã€‚ãƒ™ã‚¤ã‚ºç†è«–ã®äººæ°—ã¯æ ¹å¼·ã„ã§ã™ãã€‚ ãƒ™ãƒ¼ã‚¹ã¨ã—ã¦ã®æ•°å¼ã¯å‰²ã¨ã‚·ãƒ³ãƒ—ãƒ«ãªã®ã§ã€è‡ªåˆ†ã§å®Ÿè£…ã—ã¦ã‚‚ãã‚Œç¨‹æ‰‹é–“ã¯æŽ›ã‹ã‚‰ãªã„ã‹ã‚‚ã—ã‚Œã¾ã›ã‚“ã€‚ã—ã‹ã—ã€CPANã®ãƒ¢ã‚¸ãƒ¥ãƒ¼ãƒ«ã¨ã—ã¦æä¾›ã•ã‚Œã¦ã„ã‚‹ã®ã§ã€ãã¡ã‚‰ã‚’ä½¿ç”¨ã™ã‚‹ã®ã‚‚è‰¯ã„ã‹ã¨æ€ã„ã¾ã™ã€‚ç§ãŒçŸ¥ã£ã¦ã„ã‚‹æ‰€ã§ã¯ã€Algorithm::NaiveBayesãŒç°¡å˜ã§ä½¿ã„ã‚„ã™ã‹ã£ãŸã§ã™ã€‚ æ˜”æ›¸ã„ãŸã‚³ãƒ¼ãƒ‰ã§ã™ãŒã€ä¸‹ã®ã‚µãƒ³ãƒ—ãƒ«ã§ã¯ç°¡å˜ãªã‚¹ãƒ‘ãƒ ãƒ•ã‚£ãƒ«ã‚¿ãƒ¼ã‚’ä½œã£ã¦ã„ã¾ã™ã€‚spam.txtã¨ham.txtã¯ã€ãã‚Œãžã‚Œã®ã‚³ãƒ¼ãƒ‘ã‚¹ã‚’å½¢æ…‹ç´ è§£æžã—ã¦ä½œã£ãŸå˜èªžã®ã¿ã®ãƒªã‚¹ãƒˆã§ã™ã€‚test.txtã¯ã€åˆ¤å®šã—ãŸã„æ–‡ç« ã‹ã‚‰æŠ½å‡ºã—ãŸå˜èªžã®ãƒªã‚¹ãƒˆã§ã™ã€‚ã‚¹ãƒ‘ãƒ ã¨ãƒãƒ ã®é‡ã‚’å¢—ã‚„ã›ã°ã€ã“ã‚Œã ã‘ã§ã‚‚å‰²ã¨ä½¿ã„ç‰©ã«ãªã‚Šã¾ã™ã€‚ å¿œç”¨ä¾‹ã¨ã—ã¦ã¯ã€ã‚¹ãƒ‘ãƒ ã¨ãƒãƒ ã®2ç¨®é¡žã®ã‚«ãƒ†ã‚´ãƒªã ã‘ã§ã¯ãªãã€è¤‡æ•°ç¨®é¡žã®ã‚«ãƒ†ã‚´ãƒªã‚’ä½œã‚Œã°ãƒ–ãƒã‚°ã®
katotaku 2008/07/29
ãƒ™ã‚¤ã‚¸ã‚¢ãƒ³
ãƒªãƒ³ã‚¯
ã€Žå…¥é–€ãƒ™ã‚¤ã‚ºçµ±è¨ˆã€ã®èªã¿ã©ã“ã‚ - hiroyukikojimaâ€™s blog
ä»Šå›žã¯ã€å‰å›žã®æ—¥è¨˜ã®è£œè¶³ã€‚ å‰å›žã®çµ±è¨ˆå¦ã®é¢ç™½ã•ã¯ã©ã“ã«ã‚ã‚‹ã‹ - hiroyukikojimaã®æ—¥è¨˜ã§æ¾åŽŸæœ›å…ˆç”Ÿã®æœ¬ å…¥é–€ãƒ™ã‚¤ã‚ºçµ±è¨ˆâ€•æ„æ€æ±ºå®šã®ç†è«–ã¨ç™ºå±• ä½œè€…: æ¾åŽŸæœ›å‡ºç‰ˆç¤¾/ãƒ¡ãƒ¼ã‚«ãƒ¼: æ±äº¬å›³æ›¸ç™ºå£²æ—¥: 2008/06ãƒ¡ãƒ‡ã‚£ã‚¢: å˜è¡Œæœ¬è³¼å…¥: 107äºº ã‚¯ãƒªãƒƒã‚¯: 2,061å›žã“ã®å•†å“ã‚’å«ã‚€ãƒ–ãƒã‚° (46ä»¶) ã‚’è¦‹ã‚‹ã‚’ç´¹ä»‹ã—ãŸã€‚ãã®ã¨ãã¯ã€ã“ã®æœ¬ã‚’æ‰‹ã«ã—ã¦ã„ãªã‹ã£ãŸã®ã§ã€æ—©é€Ÿæ³¨æ–‡ã—ãŸã€‚ãã—ã¦ä»Šã€æ‰‹ã«å…¥ã£ã¦ã€ã±ã‚‰ã±ã‚‰ã¨çœºã‚ã¦ã¿ãŸã€‚ãã†ã€äºˆæƒ³é€šã‚Šã€ã“ã‚Œã¯åè‘—ã€Žçµ±è¨ˆçš„æ±ºå®šã€æ”¾é€å¤§å¦ã«å¤§å¹…åŠ ç†ã‚’ã—ãŸã‚‚ã®ã ã£ãŸã€‚ã¨ã„ã†ã‚ã‘ã§ã€ç´¹ä»‹ã—ã¦ã—ã¾ã£ãŸæ‰‹å‰ã€è²¬ä»»ã‚’ã‚‚ã£ã¦ã‚‚ã†ã¡ã‚‡ã£ã¨ãƒ•ã‚©ãƒãƒ¼ã—ãªã‘ã‚Œã°ã€ã¨æ€ã£ã¦ã“ã‚Œã‚’æ›¸ã„ã¦ã„ã‚‹ã€‚ ã“ã®æœ¬ã¯ç¢ºã‹ã«åè‘—ã§ã‚ã‚‹ã€‚ãã®ç†ç”±ã‚’ã„ãã¤ã‹æŒ™ã’ã¦ã¿ã‚ˆã†ã€‚ã¾ãšæŒ™ã’ã‚‹ã¹ãã¯ã€ ãƒ™ã‚¤ã‚ºæŽ¨å®šã®å“²å¦çš„èƒŒæ™¯ã«ã¤ã„ã¦åŒ…ã¿éš ã•ãšæ£é¢ã‹ã‚‰æ›¸ã„ã¦ã„ã‚‹ ã¨ã„ã†ç‚¹ã§ã‚ã‚‹ã€‚å‰å›žã‚‚æ›¸ã„ãŸãŒã€ãƒ™ã‚¤ã‚ºæŽ¨
katotaku 2008/07/29
ãƒ™ã‚¤ã‚¸ã‚¢ãƒ³
ãƒªãƒ³ã‚¯
AI::Categorizer
NAME AI::Categorizer - Automatic Text Categorization SYNOPSIS use AI::Categorizer; my $c = new AI::Categorizer(...parameters...); # Run a complete experiment - training on a corpus, testing on a test # set, printing a summary of results to STDOUT $c->run_experiment; # Or, run the parts of $c->run_experiment separately $c->scan_features; $c->read_training_set; $c->train; $c->evaluate_test_set; prin
katotaku 2008/07/28
ãƒ™ã‚¤ã‚¸ã‚¢ãƒ³
ãƒªãƒ³ã‚¯
ç¶±å¼•ãã«è›‡å£å½“ã¦ã‚²ãƒ¼ãƒ ï¼Ÿï¼ æ¥½ã—ãå¦ã¶ãƒ™ã‚¤ã‚ºãƒ•ã‚£ãƒ«ã‚¿ãƒ¼ã®ä»•çµ„ã¿
ä»˜ãåˆã„ãŸããªã„ã‚¹ãƒ‘ãƒ ã¨ä»˜ãåˆã†ãŸã‚ã« å—ä¿¡è€…ã®æ„å‘ã‚’ç„¡è¦–ã—ã¦ã€ä¸€æ–¹çš„ã«é€ã‚Šã¤ã‘ã‚‰ã‚Œã‚‹è¿·æƒ‘ãƒ¡ãƒ¼ãƒ«ï¼ˆã‚¹ãƒ‘ãƒ ï¼‰ã¯ã€ã„ã¾ã‚„ãƒ¡ãƒ¼ãƒ«ãƒœãƒƒã‚¯ã‚¹ã‚’é›‘éŸ³ã§ã„ã£ã±ã„ã«ã—ã¦ã—ã¾ã„ã€å¤§äº‹ãªãƒ¡ãƒ¼ãƒ«ã‚’è¦‹éŽã”ã—ã‹ããªã„ã»ã©ã®é‡ã«è†¨ã‚Œä¸ŠãŒã‚Šã€å¤§ããªå•é¡Œã¨ãªã£ã¦ã„ã¾ã™ã€‚ æ®‹å¿µãªãŒã‚‰ã€ã“ã®ã‚ˆã†ãªã‚¹ãƒ‘ãƒ ã‚’ç™ºç”Ÿæºã‹ã‚‰æ–ã¤ã‚ˆã†ãªæ ¹æœ¬çš„ãªå¯¾ç–ã¯ã„ã¾ã ã«ãªãã€ç§ãŸã¡ã¯ã€ã›ã‚ã¦ãƒ¡ãƒ¼ãƒ«ã‚µãƒ¼ãƒã§å—ã‘å–ã£ãŸå¤§é‡ã®ãƒ¡ãƒ¼ãƒ«ç¾¤ã‹ã‚‰ã‚¹ãƒ‘ãƒ ã¨å¤§äº‹ãªãƒ¡ãƒ¼ãƒ«ã‚’ä»•åˆ†ã‘ã—ã¦ãã‚Œã‚‹ä»•çµ„ã¿ã«é ¼ã‚‰ã–ã‚‹ã‚’å¾—ã¾ã›ã‚“ã€‚ ã‚¹ãƒ‘ãƒ ã‚’åˆ¤å®šã™ã‚‹æ–¹æ³•ã¯ã€æ¬¡ã®2ã¤ã«å¤§åˆ¥ã™ã‚‹ã“ã¨ãŒã§ãã¾ã™ã€‚ æœ¬ç¨¿ã§ã¯å‰è€…ã®æ–¹æ³•ã«ç€ç›®ã—ã¾ã™ã€‚ãƒ¡ãƒ¼ãƒ«ã‚’å—ã‘å–ã£ãŸäººã«ã¨ã£ã¦ã¯ã€ãƒ¡ãƒ¼ãƒ«ã®ä¸èº«ã‚’èªã‚ã°ã€ãã®ãƒ¡ãƒ¼ãƒ«ãŒã‚¹ãƒ‘ãƒ ã‹ãã†ã§ãªã„ã‹ã‚’åˆ¤å®šã™ã‚‹ã®ã¯å®¹æ˜“ãªã“ã¨ã§ã™ã€‚ã‚¹ãƒ‘ãƒ ã®å®šç¾©ã¯ã€ãƒ¡ãƒ¼ãƒ«ã‚’èªã‚€äººã«ã‚ˆã£ã¦å¤‰ã‚ã‚‹å¯èƒ½æ€§ãŒã‚ã‚Šã¾ã™ã€‚ä¾‹ãˆã°ã€ã¾ã£ãŸãã‚´ãƒ«ãƒ•ã‚’ã—ãªã„äººã«ã‚´ãƒ«ãƒ•ã®å‹§èª˜ãƒ¡ãƒ¼ãƒ«ãŒæ¥ãŸå ´åˆã¯ã‚¹ãƒ‘ãƒ ã¨ã„ãˆã‚‹ã§ã—ã‚‡ã†
katotaku 2008/03/07
ãƒ™ã‚¤ã‚¸ã‚¢ãƒ³

spam
ãƒªãƒ³ã‚¯
ã€ç¬¬4å›žã€‘ãƒ¬ã‚³ãƒ¡ãƒ³ãƒ‡ãƒ¼ã‚·ãƒ§ãƒ³ã®è™šå®Ÿï¼ˆ4ï¼‰ã€œãƒ™ã‚¤ã‚¸ã‚¢ãƒ³ã¯ã€ŒAmazonã‚’è¶…ãˆãŸã€ã®ã‹ï¼Ÿ (1/2) - ITmedia ã‚¢ãƒ³ã‚«ãƒ¼ãƒ‡ã‚¹ã‚¯
ãƒ¬ã‚³ãƒ¡ãƒ³ãƒ‡ãƒ¼ã‚·ãƒ§ãƒ³ã®è™šå®Ÿï¼ˆ4ï¼‰ï½žãƒ™ã‚¤ã‚¸ã‚¢ãƒ³ã¯ã€ŒAmazonã‚’è¶…ãˆãŸã€ã®ã‹ï¼Ÿï¼šã‚½ãƒ¼ã‚·ãƒ£ãƒ«ãƒ¡ãƒ‡ã‚£ã‚¢ ã‚»ã‚«ãƒ³ãƒ‰ã‚¹ãƒ†ãƒ¼ã‚¸ï¼ˆ1/2 ãƒšãƒ¼ã‚¸ï¼‰ Amazonã‚’è¶…ãˆã‚‹ãƒ¬ã‚³ãƒ¡ãƒ³ãƒ‡ãƒ¼ã‚·ãƒ§ãƒ³ã‚¨ãƒ³ã‚¸ãƒ³ ä»Šå¹´8æœˆ6æ—¥ãƒ»13æ—¥å·ã®ã€Žæ—¥çµŒãƒ“ã‚¸ãƒã‚¹ã€èªŒã«ã€ã€ŒçŽ‹åã¨ãƒ‹ãƒ¼ãƒˆã€€è‹¥è€…ã‚’æµªè²»ã™ã‚‹æ—¥æœ¬ç¤¾ä¼šã€ã¨ã„ã†ç‰¹é›†ãŒæŽ²è¼‰ã•ã‚ŒãŸã€‚ã“ã®ç‰¹é›†ã®ä¸ã§ç´¹ä»‹ã•ã‚Œã¦ã„ãŸã®ãŒã€ãƒ©ã‚¤ãƒ–ãƒ‰ã‚¢ã®CTOï¼ˆæœ€é«˜æŠ€è¡“è²¬ä»»è€…ï¼‰ã‚„ä»£è¡¨å–ç· å½¹ã‚’çµŒã¦ã‚¼ãƒã‚¹ã‚¿ãƒ¼ãƒˆã‚³ãƒŸãƒ¥ãƒ‹ã‚±ãƒ¼ã‚·ãƒ§ãƒ³ã‚ºã‚’è¨ç«‹ã—ãŸzakiã•ã‚“ã“ã¨å±±å´Žå¾³ä¹‹æ°ã€‚ã“ã®è¨˜äº‹ã§ã€å½¼ã®ç™»å ´ã™ã‚‹å ´é¢ã¯ãªã‚“ã¨ã‚‚å‡„ã„ã€‚æ¬¡ã®ã‚ˆã†ãªæ›¸ãå‡ºã—ã ã€‚ å¤©æ‰ãƒ—ãƒã‚°ãƒ©ãƒžãƒ¼ã®è…•ã¯ã•ã³ä»˜ã„ã¦ã„ãªã‹ã£ãŸã€‚ ã€Œã‚ˆã—ã€ã¨ã‚Šã‚ãˆãšã‚¢ãƒžã‚¾ãƒ³ã¯è¶…ãˆãŸã‹ãªã€ æ±äº¬ãƒ»æ¸‹è°·ã®å°ã•ãªã‚ªãƒ•ã‚£ã‚¹ã§ã€ã‚¼ãƒã‚¹ã‚¿ãƒ¼ãƒˆã‚³ãƒŸãƒ¥ãƒ‹ã‚±ãƒ¼ã‚·ãƒ§ãƒ³ã‚ºç¤¾é•·ã®å±±å´Žå¾³ä¹‹ã¯ã‚ãƒ¼ãƒœãƒ¼ãƒ‰ã‹ã‚‰æ‰‹ã‚’é›¢ã™ã¨ã€å°ã•ãä¼¸ã³ã‚’ã—ãŸã€‚ ã“ã®è¨˜äº‹ã®ã“ã¨ã‚’èžã„ã¦ã¿ã‚‹ã¨ã€å½¼ã¯è‹¦ç¬‘ã—ãªãŒã‚‰ã“ã†è¨€ã£ãŸã€‚ã€Œã„ã‚„ã‚ã€ã•ã™
katotaku 2007/10/02
algorithm

amazon

ãƒ™ã‚¤ã‚¸ã‚¢ãƒ³

ãƒ¬ã‚³ãƒ¡ãƒ³ãƒ‰ã‚¨ãƒ³ã‚¸ãƒ³
ãƒªãƒ³ã‚¯
Better Bayesian Filtering
January 2003 (This article was given as a talk at the 2003 Spam Conference. It describes the work I've done to improve the performance of the algorithm described in A Plan for Spam, and what I plan to do in the future.) The first discovery I'd like to present here is an algorithm for lazy evaluation of research papers. Just write whatever you want and don't cite any previous work, and indignant re
katotaku 2007/02/14
spam

ãƒ™ã‚¤ã‚¸ã‚¢ãƒ³
ãƒªãƒ³ã‚¯
Baskin School of Engineering â€“ Baskin Engineering provides unique educational opportunities, world-class research with an eye to social responsibility and diversity.
katotaku 2006/08/02
spam

ãƒ™ã‚¤ã‚¸ã‚¢ãƒ³
ãƒªãƒ³ã‚¯
A Bayesian Probability Calculus for Density Matrices
Be bold Inspiring engineers to elevate the world through bold and socially responsible innovation Join Us Be boundless Inspiring engineers to elevate the world through bold and socially responsible innovation Join Us Be a Baskin Engineer Inspiring engineers to elevate the world through bold and socially responsible innovation Join Us
katotaku 2006/08/02
spam

ãƒ™ã‚¤ã‚¸ã‚¢ãƒ³
ãƒªãƒ³ã‚¯
Filter::POPFileã§è¤‡æ•°ã®ãƒ™ã‚¤ã‚ºè¾žæ›¸ - Charsbar::Note
å®Ÿè£…ã™ã‚‹ã¨ã—ãŸã‚‰ã€å„ãƒ•ã‚£ãƒ¼ãƒ‰ã”ã¨ã«ãƒ™ã‚¤ã‚ºç†è«–ã§å¦ç¿’ã•ã›ãªã„ã¨ã‚ã‚“ã¾ã†ã‚Œã—ããªã•ãã†ã€‚A ã•ã‚“ãŒæ›¸ãç§ç”Ÿæ´»ãƒã‚¿ã¯ã¤ã¾ã‚‰ãªã„ã‹ã‚‰èªã¿ãŸããªã„ã‘ã©ã€B ã•ã‚“ãŒæ›¸ãç§ç”Ÿæ´»ãƒã‚¿ã¯ãŠã‚‚ã‚ã„ã‹ã‚‰èªã¿ãŸã„ã€ã¨ãã«åˆ¤æ–ã™ã‚‹ã®ãŒã²ã¨ã¤ã®å¦ç¿’ç©ºé–“ã—ã‹ãªã„ã¨ç„¡ç†ãã’ã€‚ http://subtech.g.hatena.ne.jp/secondlife/20060514/1147541796 ãƒ¡ãƒ¼ãƒ«ã®ãƒ™ã‚¤ã‚¸ã‚¢ãƒ³ãƒ•ã‚£ãƒ«ã‚¿ã®å ´åˆã€Header ã§é‡ã¿ä»˜ã‘ã‚„ã‚‰åˆ†é¡žã¯ã§ãã‚‹ã®ã¯çŸ¥ã£ã¦ã‚‹ã‚“ã§ã™ãŒã€ãƒ™ã‚¤ã‚ºã§å¦ç¿’ã™ã‚‹ã¨ã“ã¯ä¸€ã‚«æ‰€ã ã‘ã ã¨æ€ã£ã¦ãŸã‚“ã§ã™ãŒã€ç™ºä¿¡å…ƒã”ã¨ã«å¦ç¿’ã™ã‚‹ç®‡æ‰€ã‚‚åˆ¥ç©ºé–“ã§è¡Œãˆã‚‹ã‚“ã§ã—ã‚‡ã†ã‹ãï¼Ÿ http://subtech.g.hatena.ne.jp/secondlife/20060514/1147582785 ä¸Šã®è©±ã«ã¤ã„ã¦ã¯ãƒ–ã‚¯ãƒžã‚³ãƒ¡ãƒ³ãƒˆã§ã‚‚ã€Œãªã‚“ã§ã‚‚ãƒ™ã‚¤ã‚ºã ã‘ã§è§£æ±ºã—ã‚ˆã†ã¨ã„ã†ã®ã¯ç„¡ç†ãŒã‚ã‚‹ã€‚ãã®è¾ºã¯POP
katotaku 2006/05/19
spam

ãƒ™ã‚¤ã‚¸ã‚¢ãƒ³
ãƒªãƒ³ã‚¯
Better Bayesian Filtering - ãƒ™ã‚¤ã‚¸ã‚¢ãƒ³ãƒ•ã‚£ãƒ«ã‚¿ã®æ”¹å–„
ãƒ™ã‚¤ã‚¸ã‚¢ãƒ³ãƒ•ã‚£ãƒ«ã‚¿ã®æ”¹å–„ --- Better Bayesian Filtering Paul Graham, January 2003 ã“ã‚Œã¯ã€Paul Grahamï¼š Better Bayesian Filtering ã‚’ã€åŽŸè‘—è€…ã®è¨±å¯ã‚’å¾—ã¦ç¿»è¨³ãƒ»å…¬é–‹ã™ã‚‹ã‚‚ã®ã§ã™ã€‚ <ç‰ˆæ¨©è¡¨ç¤º> æœ¬å’Œè¨³ãƒ†ã‚ã‚¹ãƒˆã®è¤‡è£½ã€å¤‰æ›´ã€å†é…å¸ƒã¯ã€ã“ã®ç‰ˆæ¨©è¡¨ç¤ºã‚’æ®‹ã™é™ã‚Šã€è‡ªç”±ã«è¡Œã£ã¦çµæ§‹ã§ã™ã€‚ (ã€Œã“ã®ç‰ˆæ¨©è¡¨ç¤ºã€ã«ã¯ä¸Šã®æ–‡ã‚‚å«ã¾ã‚Œã¾ã™ã€‚ã™ãªã‚ã¡ã€å†é…å¸ƒã‚’ç¦æ¢ã—ã¦ã¯ã„ã‘ã¾ã›ã‚“)ã€‚ Copyright 2002 by Paul Graham åŽŸæ–‡: http://www.paulgraham.com/better.html æ—¥æœ¬èªžè¨³ï¼šShiro Kawai (shiro ï¼ acm.org) <ç‰ˆæ¨©è¡¨ç¤ºçµ‚ã‚Š> Paul Grahamæ°ã®ã‚¨ãƒƒã‚»ã‚¤ã‚’ã¾ã¨ã‚ãŸã€Žãƒãƒƒã‚«ãƒ¼ã¨ç”»å®¶ã€ã® é‚¦è¨³ç‰ˆãŒå‡ºç‰ˆã•ã‚Œã¾ã—ãŸã€‚ å‡º
katotaku 2006/05/06
mail

spam

ãƒ™ã‚¤ã‚¸ã‚¢ãƒ³
ãƒªãƒ³ã‚¯
æ¬§ç¾Žäººä¸Žç¦½ZOZ0æ€§ä¼¦äº¤,97äººå¦»äººäººåšäººç¢°äººäººçˆ½,ä¹…ä¹…ä¹…ç²¾å“äººå¦»ä¸€åŒºäºŒåŒºä¸‰åŒºèœœæ¡ƒ,å¼ºå¥·ç§˜ä¹¦å¸ä¹³å…è´¹è§‚çœ‹
æ¬§ç¾Žäººä¸Žç¦½ZOZ0æ€§ä¼¦äº¤,97äººå¦»äººäººåšäººç¢°äººäººçˆ½,ä¹…ä¹…ä¹…ç²¾å“äººå¦»ä¸€åŒºäºŒåŒºä¸‰åŒºèœœæ¡ƒ,å¼ºå¥·ç§˜ä¹¦å¸ä¹³å…è´¹è§‚çœ‹,æˆ äºº é»„ ç‰‡å…è´¹è§‚çœ‹,ä¸€äºŒä¸‰å››æ—¥æœ¬é«˜æ¸…è§†é¢‘åŠ¨æ¼«,è‰²ç»¼åˆä¹…ä¹…ä¹…ä¹…ä¹…ä¹…ç»¼åˆ
katotaku 2006/05/06
perl

mail

ãƒ™ã‚¤ã‚¸ã‚¢ãƒ³
ãƒªãƒ³ã‚¯
Algorithm::NaiveBayesã®ç·´ç¿’ - [çµ] 2004å¹´4æœˆ - çµåŸŽæµ©ã®æ—¥è¨˜
ç›®æ¬¡ 2004å¹´4æœˆ30æ—¥ - ã¯ã¡ã“ã•ã‚“ã€ãŠç–²ã‚Œã•ã¾ / ã€ŽçµåŸŽæµ©ã®Wikiå…¥é–€ã€ã¸ã®ãŸã ã•ã‚“ã®æ›¸è©• / æ—¥è¨˜ã¸ã®ãƒªãƒ•ã‚¡ãƒ© / 2004å¹´4æœˆ29æ—¥ - tDiaryã‚µãƒ³ãƒ—ãƒ©ãƒ¼ç‰ˆå£°ã®ã‹ã‘ã‚‰ã€‚ / å£°ã®ã‹ã‘ã‚‰ã€‚ã®æ¬¡ã€‚ / è‰¯ã„spamãƒ•ã‚£ãƒ«ã‚¿ã¯ã‚ã‚‹ã‹ã—ã‚‰ã‚“ / 2004å¹´4æœˆ26æ—¥ - å£°ã®ã‹ã‘ã‚‰ã€‚0.9.0ãƒªãƒªãƒ¼ã‚¹ / 2004å¹´4æœˆ24æ—¥ - ãŠã¯ã‚ˆã†ã”ã–ã„ã¾ã™ / 2004å¹´4æœˆ23æ—¥ - ã€ŽçµåŸŽæµ©ã®Wikiå…¥é–€ã€ãƒ©ãƒ³ã‚ãƒ³ã‚° / 2004å¹´4æœˆ22æ—¥ - ç§‘å¦ã®æ–¹æ³• / 2004å¹´4æœˆ21æ—¥ - HTMLé¢¨ã®URLã§CGIã‚’å‹•ã‹ã™ãŸã‚ã®ãƒ¡ãƒ¢ / 2004å¹´4æœˆ20æ—¥ - 4æ³ã®æ¬¡ç”·ãŒç”Ÿå”ã®ç®±ã‚’é‹ã¶ / 2004å¹´4æœˆ18æ—¥ - Algorithm::NaiveBayesã®ç·´ç¿’ / ãŠã¯ã‚ˆã†ã”ã–ã„ã¾ã™ / 2004å¹´4æœˆ17æ—¥ - Jcodeã§ã‚¨ãƒ©ãƒ¼â†’CGI::CarpãŒå¤ã‹ã£ãŸ /
katotaku 2006/05/06
perl

spam

ãƒ™ã‚¤ã‚¸ã‚¢ãƒ³
ãƒªãƒ³ã‚¯
Algorithm::NaiveBayes
NAME Algorithm::NaiveBayes - Bayesian prediction of categories SYNOPSIS use Algorithm::NaiveBayes; my $nb = Algorithm::NaiveBayes->new; $nb->add_instance (attributes => {foo => 1, bar => 1, baz => 3}, label => 'sports'); $nb->add_instance (attributes => {foo => 2, blurp => 1}, label => ['sports', 'finance']); ... repeat for several more instances, then: $nb->train; # Find results for unseen instan
katotaku 2006/05/06
perl

ãƒ™ã‚¤ã‚¸ã‚¢ãƒ³
ãƒªãƒ³ã‚¯
A Plan for Spam
August 2002 (This article describes the spam-filtering techniques used in the spamproof web-based mail reader we built to exercise Arc. An improved algorithm is described in Better Bayesian Filtering.) I think it's possible to stop spam, and that content-based filters are the way to do it. The Achilles heel of the spammers is their message. They can circumvent any other barrier you set up. They ha
katotaku 2006/05/06
mail

spam

ãƒ™ã‚¤ã‚¸ã‚¢ãƒ³
ãƒªãƒ³ã‚¯
çª“ã®æœ - ã€NEWSã€‘ISO-2022-JPä»¥å¤–ã®æ—¥æœ¬èªžãƒ¡ãƒ¼ãƒ«ã‚‚æŒ¯ã‚Šåˆ†ã‘å¯èƒ½ã«ãªã£ãŸã€ŒPOPFileã€v0.22.0
â€œãƒ™ã‚¤ã‚ºç†è«–â€ã‚’ã‚‚ã¨ã«è¿·æƒ‘ãƒ¡ãƒ¼ãƒ«å¯¾ç–ãŒã§ãã‚‹POPãƒ—ãƒã‚ã‚·ãƒ¼ã‚µãƒ¼ãƒãƒ¼ã€ŒPOPFileã€ãŒ8æ—¥ã€v0.22.0ã«ãƒãƒ¼ã‚¸ãƒ§ãƒ³ã‚¢ãƒƒãƒ—ã—ã€ISO-2022-JPä»¥å¤–ã®æ–‡å—ã‚³ãƒ¼ãƒ‰ã§è¨˜è¿°ã•ã‚ŒãŸæ—¥æœ¬èªžãƒ¡ãƒ¼ãƒ«ã‚‚æŒ¯ã‚Šåˆ†ã‘å¯èƒ½ã«ãªã£ãŸã€‚ã¾ãŸã€ã“ã‚Œã¾ã§æ—¥æœ¬èªžãƒ¡ãƒ¼ãƒ«ã®ä»¶åã¯1ã¤ã®æ–‡å—åˆ—ã¨ã—ã¦æ‰±ã‚ã‚Œã¦ã„ãŸãŒã€æœ¬ãƒãƒ¼ã‚¸ãƒ§ãƒ³ã‹ã‚‰ã¯ãƒ¡ãƒ¼ãƒ«æœ¬æ–‡ã¨åŒæ§˜ã«ã€å˜èªžã§åŒºåˆ‡ã£ã¦å‡¦ç†ã™ã‚‹ã‚ˆã†ã«ä»•æ§˜ãŒå¤‰æ›´ã•ã‚Œã¦ã„ã‚‹ã€‚ ã•ã‚‰ã«ã€æŒ¯ã‚Šåˆ†ã‘æ¸ˆã¿ã®ãƒ¡ãƒ¼ãƒ«ã‚’ç®¡ç†ã™ã‚‹ï¼»å±¥æ´ï¼½ã‚¿ãƒ–ã§ã€å¾“æ¥ã¯å·®å‡ºäººãƒ»ä»¶åãƒ»æŒ¯ã‚Šåˆ†ã‘çŠ¶æ…‹ã®3é …ç›®ã—ã‹ç¢ºèªã§ããªã‹ã£ãŸãŒã€å—ä¿¡æ—¥ãƒ»å·®å‡ºäººãƒ»å®›å…ˆãƒ»ä»¶åãªã©åˆè¨ˆ8é …ç›®ã‚’è‡ªç”±ã«ã‚«ã‚¹ã‚¿ãƒžã‚¤ã‚ºã—ã¦è¡¨ç¤ºå¯èƒ½ã«ãªã£ãŸã€‚ã¾ãŸã€ï¼»å±¥æ´ï¼½ã‚¿ãƒ–ã®å·®å‡ºäººã‚„ä»¶åæŒ‡å®šã«ã‚ˆã‚‹ãƒ¡ãƒ¼ãƒ«æ¤œç´¢æ©Ÿèƒ½ã§ã€æ¡ä»¶ã«ä¸€è‡´ã™ã‚‹ã‚‚ã®ä»¥å¤–ã«ä¸€è‡´ã—ãªã„ã‚‚ã®ã‚‚æ¤œç´¢ã§ãã‚‹ã‚ˆã†ã«ãªã£ã¦ã„ã‚‹ã€‚ ãã®ã»ã‹ã€APOP/SSLèªè¨¼ã¸ã®å¯¾å¿œã€NNTPãƒ—ãƒã‚ã‚·ãƒ¼ã‚µãƒ¼ãƒãƒ¼æ©Ÿèƒ½ã®è¿½åŠ ã€ä»–ã‚½ãƒ•ãƒˆã§æœ¬ã‚½ãƒ•ãƒˆ
katotaku 2005/12/17
ãƒ™ã‚¤ã‚¸ã‚¢ãƒ³
ãƒªãƒ³ã‚¯
andore.com
This domain may be for sale!
katotaku 2005/10/27
ãƒ™ã‚¤ã‚¸ã‚¢ãƒ³
ãƒªãƒ³ã‚¯
bsfilter / bayesian spam filter / ãƒ™ã‚¤ã‚¸ã‚¢ãƒ³ ã‚¹ãƒ‘ãƒ ãƒ•ã‚£ãƒ«ã‚¿
Subscribe to our newsletter to get daily updates about Programming and Spam Filtering.
katotaku 2005/10/04
ãƒ™ã‚¤ã‚¸ã‚¢ãƒ³
ãƒªãƒ³ã‚¯
Microsoft PowerPoint - 031220-SPAM.ppt
1 Bayesian Filteringã‚’ç”¨ã„ãŸ spamå¯¾ç– æ˜Žæ²»å¤§å¦æƒ…å ±ã‚·ã‚¹ãƒ†ãƒ ç®¡ç†èª² æœéƒ¨è£•ä¹‹ [email protected] 2 spamã ã‚‰ã‘ã®æœ å•†å“æ¡ˆå†… å•†å“æ¡ˆå†… ããšã¿è¬›ã¾ãŒã„ ããšã¿è¬›ã¾ãŒã„ ã‚¯ã‚¹ãƒª ã‚¯ã‚¹ãƒª ã‚¨ãƒ ã‚¨ãƒ åç°¿å±‹ åç°¿å±‹ æœã®ä»•äº‹ã¯ã€ã¾ãšã€ãƒ¡ãƒ¼ãƒ«ã‚’æ¶ˆã™ã“ã¨ã‹ã‚‰ 2 3 æœ¬æ—¥ã®ãƒ¡ãƒ‹ãƒ¥ãƒ¼ { spamã‚’æŽ’é™¤ã™ã‚‹æ–¹æ³•ã¯ï¼Ÿ { ä»Šã€æ³¨ç›®ã®ã€ŒBayesian Filteringã€ã¨ã¯ï¼Ÿ { ä½¿ã„ã‚‚ã®ã«ãªã‚‹ã®ã‹ï¼Ÿ { GraceM ailã‹ã‚‰ã®åˆ©ç”¨ã¯å¯èƒ½ã‹ï¼Ÿ 4 spamã®æŽ’é™¤æ–¹æ³• ãƒ»ãƒ»ãƒ»ãƒ»ãƒ»ãƒ» ãƒ»ãƒ»ãƒ»ãƒ» ãƒ»ãƒ»ãƒ»ãƒ»ãƒ» ãƒ¡ãƒ¼ãƒ«ã‚µãƒ¼ãƒ ã§ãƒ•ã‚£ãƒ«ã‚¿ ãƒ¡ãƒ¼ãƒ«ã‚¯ãƒ©ã‚¤ã‚¢ãƒ³ãƒˆ ã§ãƒ•ã‚£ãƒ«ã‚¿ åˆ¤åˆ¥å ´æ‰€ï¼‘ åˆ¤åˆ¥å ´æ‰€ï¼‘ åˆ¤åˆ¥å ´æ‰€ï¼’ åˆ¤åˆ¥å ´æ‰€ï¼’ 3 5 spamå¯¾ç–æŠ€è¡“ spamå¯¾ç– ä¸ç¶™ã—ã¦ã„ã‚‹ ãƒ¡ãƒ¼ãƒ«ã‚µãƒ¼ãƒ ã«æ³¨ç›® ãƒ¡ãƒ¼ãƒ«ã®ä¸èº« ï¼ˆãƒ˜ãƒƒãƒ€ã€ãƒœãƒ‡ã‚£ï¼‰ ã«æ³¨ç›® OpenRelay ãƒ‡ãƒ¼
katotaku 2005/10/04
ãƒ™ã‚¤ã‚¸ã‚¢ãƒ³
ãƒªãƒ³ã‚¯
http://hotwired.goo.ne.jp/webmonkey/2003/09/index2a_page5.html
katotaku 2005/10/04
ãƒ™ã‚¤ã‚¸ã‚¢ãƒ³
ãƒªãƒ³ã‚¯
1 2 æ¬¡ã®ãƒšãƒ¼ã‚¸