[B! algorithm] kiririmodeã®ãƒ–ãƒƒã‚¯ãƒžãƒ¼ã‚¯

kiririmode id:kiririmode

algorithmã«é–¢ã™ã‚‹kiririmodeã®ãƒ–ãƒƒã‚¯ãƒžãƒ¼ã‚¯ (27)

${{author_name}}$

{{{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}}

OpenSearch ã«ãŠã‘ã‚‹ 10 å„„è¦æ¨¡ã®ãƒ¦ãƒ¼ã‚¹ã‚±ãƒ¼ã‚¹ã«é©ã—ãŸ k-NN ã‚¢ãƒ«ã‚´ãƒªã‚ºãƒ ã®é¸å®š | Amazon Web Services
Amazon Web Services ãƒ–ãƒã‚° OpenSearch ã«ãŠã‘ã‚‹ 10 å„„è¦æ¨¡ã®ãƒ¦ãƒ¼ã‚¹ã‚±ãƒ¼ã‚¹ã«é©ã—ãŸ k-NN ã‚¢ãƒ«ã‚´ãƒªã‚ºãƒ ã®é¸å®š ã“ã®è¨˜äº‹ã¯ã€Choose the k-NN algorithm for your billion-scale use case with OpenSearch ã‚’ç¿»è¨³ã—ãŸã‚‚ã®ã§ã™ã€‚ è‡ªç„¶è¨€èªžå‡¦ç† (NLP) ã‚·ã‚¹ãƒ†ãƒ ã‚„ã€ãƒ¬ã‚³ãƒ¡ãƒ³ãƒ‰ã‚¨ãƒ³ã‚¸ãƒ³ã€ã¾ãŸã¯æ¤œç´¢ã‚·ã‚¹ãƒ†ãƒ ãªã©ã®æ©Ÿæ¢°å¦ç¿’ (ML) ã‚¢ãƒ—ãƒªã‚±ãƒ¼ã‚·ãƒ§ãƒ³ã‚’æ§‹ç¯‰ã—ã‚ˆã†ã¨ã™ã‚‹ã¨ã€ãƒ¯ãƒ¼ã‚¯ãƒ•ãƒãƒ¼ã®ã©ã“ã‹ã§ k æœ€è¿‘å‚ (k-NN) æ¤œç´¢ã‚¢ãƒ«ã‚´ãƒªã‚ºãƒ ã‚’ä½¿ç”¨ã™ã‚‹ã“ã¨ãŒã‚ˆãã‚ã‚Šã¾ã™ã€‚ãƒ‡ãƒ¼ã‚¿ãƒã‚¤ãƒ³ãƒˆã®æ•°ãŒæ•°å„„ã‚„æ•°åå„„ã«é”ã™ã‚‹ã¨ã€k-NN æ¤œç´¢ã‚·ã‚¹ãƒ†ãƒ ã®ã‚¹ã‚±ãƒ¼ãƒªãƒ³ã‚°ãŒå¤§ããªèª²é¡Œã¨ãªã‚Šã¾ã™ã€‚è¿‘ä¼¼æœ€è¿‘å‚ (ANN) æ¤œç´¢ã‚¢ãƒ«ã‚´ãƒªã‚ºãƒ ã¯ã€ã“ã®èª²é¡Œã‚’è§£æ±ºã™ã‚‹ãŸã‚ã®å„ªã‚ŒãŸæ–¹æ³•ã® 1 ã¤ã§ã™ã€‚ k-NN ã¯ã€ä»–ã® ML ã®æŠ€è¡“ã¨æ¯”
kiririmode 2024/09/07
è¿‘ä¼¼æœ€è¿‘å‚æ¤œç´¢ã«é–¢ã™ã‚‹HNSWã€IVF+PQã®èª¬æ˜Žã¨æ¯”è¼ƒã€‚IVFã¯ãƒ™ã‚¯ãƒˆãƒ«ã‚’ã‚¯ãƒ©ã‚¹ã‚¿ãƒªãƒ³ã‚°ã—ã¦ã‚¯ãƒ©ã‚¹ã‚¿å†…ã®ã‚µãƒ–ã‚»ãƒƒãƒˆã‚’æ¤œç´¢ã€‚PQã¯ãƒ™ã‚¯ãƒˆãƒ«ã‚’é‡ååŒ–ã™ã‚‹ã“ã¨ã§æƒ…å ±é‡ã‚’å‰Šæ¸›

opensearch

ai

algorithm
ãƒªãƒ³ã‚¯
Hierarchical Navigable Small Worlds (HNSW) | Pinecone
Hierarchical Navigable Small World (HNSW) graphs are among the top-performing indexes for vector similarity search[1]. HNSW is a hugely popular techno logy that time and time again produces state-of-the-art performance with super fast search speeds and fantastic recall. Yet despite being a popular and robust algorithm for approximate nearest neighbors (ANN) searches, understanding how it works is f
kiririmode 2024/09/07
ãƒ™ã‚¯ãƒˆãƒ«ã®è¿‘å‚æ¤œç´¢ã§ç”¨ã„ã‚‰ã‚Œã‚‹HNSWã®æ§‹ç¯‰/æ¤œç´¢ã‚¢ãƒ«ã‚´ãƒªã‚ºãƒ ã€‚Mã‚„efsearchã€efconstructionã‚’å¤§ããã™ã‚Œã°recallã¯å¤§ãããªã‚‹ãŒæ§‹ç¯‰/æ¤œç´¢æ™‚é–“ã‚„ãƒ¡ãƒ¢ãƒªä½¿ç”¨é‡ãŒãƒˆãƒ¬ãƒ¼ãƒ‰ã‚ªãƒ•

ai

datastructure

algorithm
ãƒªãƒ³ã‚¯
https://arxiv.org/pdf/2212.10496.pdf
kiririmode 2023/11/12
HyDEã®è«–æ–‡ã€‚HyDEã§ã¯ã€è³ªå•ã«å¯¾ã—ã¦LLMã§ä»®æƒ³ãƒ‰ã‚ãƒ¥ãƒ¡ãƒ³ãƒˆã‚’ç”Ÿæˆã—ãŸä¸Šã§ã€ãã®æ–‡æ›¸ã¨ãƒŠãƒ¬ãƒƒã‚¸ãƒ™ãƒ¼ã‚¹ã¨ã®é–¢é€£ã‚’è¨ˆç®—ã™ã‚‹å½¢ã§retrievalã‚’è¡Œã†

rag

ai

algorithm
ãƒªãƒ³ã‚¯
RAGã«ãŠã‘ã‚‹ãƒ‰ã‚ãƒ¥ãƒ¡ãƒ³ãƒˆæ¤œç´¢ç²¾åº¦å‘ä¸Šã«ã¤ã„ã¦(æ¦‚è¦ç·¨)
ã¯ã˜ã‚ã¾ã—ã¦ã€‚æå®³ä¿é™ºã‚¸ãƒ£ãƒ‘ãƒ³æ ªå¼ä¼šç¤¾ DXæŽ¨é€²éƒ¨ã®çœžæ–¹ã§ã™ã€‚æ™®æ®µã¯ãƒªãƒ¼ãƒ‰ã‚¨ãƒ³ã‚¸ãƒ‹ã‚¢ã¨ã—ã¦ã€æ–°ã—ã„ã‚µãƒ¼ãƒ“ã‚¹ã®ã‚¢ãƒ¼ã‚ãƒ†ã‚¯ãƒãƒ£æ¤œè¨Žã‹ã‚‰ãƒãƒ¼ãƒ³ãƒã¾ã§ã®ä½œæ¥ã‚„ã€æ–°è¦æŠ€è¡“ã‚’ç”¨ã„ãŸã‚¢ãƒ—ãƒªã®ãƒ—ãƒãƒˆã‚¿ã‚¤ãƒ—å®Ÿè£…ãªã©ã‚’è¡Œãªã£ã¦ã„ã¾ã™ã€‚ å¼Šç¤¾ã§ã¯ã€LLMï¼ˆLarge Language Modelsï¼‰ã‚’æ´»ç”¨ã—ãŸã‚¢ãƒ—ãƒªã‚±ãƒ¼ã‚·ãƒ§ãƒ³ã®é–‹ç™ºã‚’ç©æ¥µçš„ã«æ¤œè¨Žã—ã€æ—¢ã«ç¤¾å†…ã§ã„ãã¤ã‹ã®ãƒ—ãƒãƒˆã‚¿ã‚¤ãƒ—ã‚’ãƒãƒ¼ãƒ³ãƒã—ã¦ã„ã¾ã™ã€‚ æœ¬è¨˜äº‹ã§ã¯ã€ãã®æœ€ã‚‚ä¸€èˆ¬çš„ï¼Ÿãªãƒ¦ãƒ¼ã‚¹ã‚±ãƒ¼ã‚¹ã®ä¸€ã¤ã¨ã‚‚è¨€ãˆã‚‹RAGï¼ˆRetrieval Augmented Generativeï¼‰ã®æ§‹ç¯‰ã«ãŠã„ã¦ã€ãƒ‰ã‚ãƒ¥ãƒ¡ãƒ³ãƒˆæ¤œç´¢ç²¾åº¦ã®å‘ä¸Šã«ã©ã®ã‚ˆã†ã«å–ã‚Šçµ„ã‚“ã å†…å®¹ã®æ¦‚è¦ã‚’ç´¹ä»‹ã•ã›ã¦ã„ãŸã ãã¾ã™ã€‚å®Ÿéš›ã®è©³ç´°ãªæ‰‹æ³•ãŠã‚ˆã³çµæžœã«ã¤ã„ã¦ã¯ã€åˆ¥è¨˜äº‹ï¼ˆå®Ÿè·µç·¨ï¼‰ã§è§£èª¬äºˆå®šã§ã™ã€‚ ã¯ã˜ã‚ã« RAGã¨ã¯ï¼Ÿ ã“ã®è¨˜äº‹ã‚’èªã¾ã‚Œã¦ã„ã‚‹æ–¹ã®ä¸ã«ã¯ã”å˜çŸ¥ã®æ–¹ã‚‚å¤šã„ã§ã—ã‚‡ã†ãŒã€RAGã¨ã¯Retrieval A
kiririmode 2023/11/12
RAGã§ã®ãƒ‰ã‚ãƒ¥ãƒ¡ãƒ³ãƒˆæ¤œç´¢ç²¾åº¦ã‚’å‘ä¸Šã™ã‚‹ãŸã‚ã®ã•ã¾ã–ã¾ãªæ‰‹æ³•ã®ç´¹ä»‹ã€‚å¯¾è±¡ãƒ‰ã‚ãƒ¥ãƒ¡ãƒ³ãƒˆã‚’ã‚µãƒžãƒªã—ã¦ãŠã„ãŸã‚Šã€è³ªå•ã®æ–¹ã‚’æ‹¡å¼µã—ãŸã‚Š(HyDEå«ã‚€ï¼‰ã€é–¢é€£æ–‡æ›¸ã‚’re-rankingã—ãŸã‚Šã€‚

ai

rag

algorithm
ãƒªãƒ³ã‚¯
BM25ã‚’æ•°å¼ã‹ã‚‰èª¬æ˜Žã™ã‚‹ - Qiita
ã¯ã˜ã‚ã« BM25ã¯ç‰¹ã«æ¤œç´¢ã‚¢ãƒ«ã‚´ãƒªã‚ºãƒ ã«ä½¿ã‚ã‚Œã‚‹è‡ªç„¶è¨€èªžå‡¦ç†ã®ä¸€ã¤ã§ã€tf-idfã®é€²åŒ–ç³»ã§ã‚ã‚‹ã€‚å…·ä½“çš„ã«ã¯å˜èªžã®å‡ºç¾é »åº¦ã«åŸºã¥ã„ã¦ã€æ–‡ç« ã®é †ä½ä»˜ã‘ã‚’è¡Œã†ã€‚tf-idfã¨ã®é•ã„ã¯ãƒ‰ã‚ãƒ¥ãƒ¡ãƒ³ãƒˆãŒçŸã„ã»ã©é †ä½ãŒé«˜ãã¤ãã€é•·ã„ã»ã©é †ä½ãŒä½Žãã¤ãå‚¾å‘ãŒã‚ã‚‹ã¨ã„ã†ã¨ã“ã‚ã§ã‚ã‚‹ã€‚ã“ã®è¨˜äº‹ã§ã¯æ•°å¼ã‚’ç´ã©ã„ã¦ã€BM25ã®æ€§è³ªã‚’èª¬æ˜Žã™ã‚‹ã€‚ æ•°å¼ BM25ã®æ•°å¼ã«ã¤ã„ã¦ã¾ãšèª¬æ˜Žã™ã‚‹ã€‚$D$ã‚’æ–‡ç« å…¨ä½“ã®é›†åˆ(ä»¥ä¸‹å…¨æ–‡ç« ã¨å‘¼ã¶)ã€$d$ã¯æ–‡ç« ã§ã‚ã‚Š$D$ã®è¦ç´ ã€$q$ã‚’æ¤œç´¢ã‚¯ã‚¨ãƒª($q_i\in q$)ã¨ã—ãŸæ™‚ã®BM25ã®æ•°å¼ã¯ä»¥ä¸‹ã®ã‚ˆã†ãªã‚‚ã®ã§ã‚ã‚‹ã€‚ $$ score(q, d) = \sum_i idf(q_i)\times\frac{(k_1+1)f(q_i, d)}{f(q_i, d)+k_1(1-b+b\frac{|d|}{avg(dl)})} $$ $idf(q_i)$ã¯å˜èªž$q_i$ã®idfã€$f
kiririmode 2023/11/12
å˜èªžã®å‡ºç¾é »åº¦ã«åŸºã¥ã„ã¦æ–‡ç« é–“ã®é–¢é€£ã®å¼·ã•ã‚’ãƒ©ãƒ³ã‚¯ã¥ã‘ã™ã‚‹

ai

algorithm
ãƒªãƒ³ã‚¯
GitHub - erikbern/ann-benchmarks: Benchmarks of approximate nearest neighbor libraries in Python
You signed in with another tab or window. Reload to refresh your session. You signed out in another tab or window. Reload to refresh your session. You switched accounts on another tab or window. Reload to refresh your session. Dismiss alert
kiririmode 2023/11/08
ãƒ™ã‚¯ãƒˆãƒ«æ¤œç´¢ã®ã‚¢ãƒ«ã‚´ãƒªã‚ºãƒ ANNã®ãƒ‘ãƒ•ã‚©ãƒ¼ãƒžãƒ³ã‚¹æ¯”è¼ƒ

ai

vector

algorithm
ãƒªãƒ³ã‚¯
GeoTools ã§åœ°çƒåœ°å›³èµ·æºã®æµ·å²¸ç·šãƒ‡ãƒ¼ã‚¿ã‚’å˜ç´”åŒ–ã™ã‚‹ã€€ãã®ï¼“ - Relevant, Timely, and Accurate
d:id:hfu:20071211, d:id:hfu:20071212 ã®ç¶šãã¨ã—ã¦ã€ãƒãƒªã‚´ãƒ³ã®å˜ç´”åŒ–ã‚’å®Ÿéš›ã«ã‚„ã£ã¦ã¿ã¾ã—ãŸã€‚ geotools.rb ã®æ”¹å¤‰ geotools.rb ã‚’æ”¹å¤‰ã—ã€ãƒãƒªã‚´ãƒ³ã®å˜ç´”åŒ–ã«å¿…è¦ãªã‚¯ãƒ©ã‚¹ã‚’å–ã‚Šè¾¼ã‚€ã‚ˆã†ã«ã—ã¾ã—ãŸã€‚ã“ã®ã‚¨ãƒ³ãƒˆãƒªã®ã‚¹ã‚¯ãƒªãƒ—ãƒˆã‚’å®Ÿè¡Œã™ã‚‹ã«ã¯ã€http://svg mapdata.sakura.ne.jp/geotools/ ã«ã‚ã‚‹ geotools.rb (Time-stamp: <2007-12-13 05 : 47 : 35 hfu> ä»¥é™)ã‚’ä½¿ã£ã¦ãã ã•ã„ã€‚ ãƒãƒªã‚´ãƒ³ã®å˜ç´”åŒ–ã«å¿…è¦ãªã‚¯ãƒ©ã‚¹ã¨ã¯ ãƒãƒªã‚´ãƒ³ã®å˜ç´”åŒ–ã‚’ã—ã¦ãã‚Œã‚‹ã‚¯ãƒ©ã‚¹ã«ã¯ä»¥ä¸‹ã®ï¼’ã¤ãŒã‚ã‚Šã¾ã™ï¼š com.vividsolutions.jts.simplify.DouglasPeuckerSimplifier åœ°å›³å¦çš„ãªãƒ™ã‚¯ãƒˆãƒ«ãƒ‡ãƒ¼ã‚¿ã®å˜ç´”åŒ–ã‚¢ãƒ«ã‚´ãƒªã‚ºãƒ ã¨ã—ã¦ã¯å¤å…¸çš„ã§ã‚ã‚‹ã‚‰ã—ã„
kiririmode 2023/06/03
Douglas-Peuckerã‚¢ãƒ«ã‚´ãƒªã‚ºãƒ ã§é ‚ç‚¹æ•°ã‚’æ¸›ã‚‰ã—ãŸä¾‹ã€‚è‰¯è³ªãªçµæžœ

algorithm

snow
ãƒªãƒ³ã‚¯
Simplify â€” GeoTools 31-SNAPSHOT User Guide
kiririmode 2023/06/03
Douglas-Peuckerã‚¢ãƒ«ã‚´ãƒªã‚ºãƒ ã¯GeoToolsãƒ©ã‚¤ãƒ–ãƒ©ãƒªã§å®Ÿè£…ã•ã‚Œã¦ã„ã‚‹

java

algorithm

snow
ãƒªãƒ³ã‚¯
ãƒ©ã‚¤ãƒ³ã®å˜ç´”åŒ–
kiririmode 2023/06/03
è»Šä¸¡è»Œè·¡ãƒ‡ãƒ¼ã‚¿ã®ç‚¹ãƒ‡ãƒ¼ã‚¿ãŒå¤šã„å ´åˆã€é‡è¦ã§ãªã„ç‚¹ãƒ‡ãƒ¼ã‚¿ã‚’çœç•¥ã™ã‚‹ã‚¢ãƒ«ã‚´ãƒªã‚ºãƒ ã¨ã—ã¦Douglas-Peukerã‚¢ãƒ«ã‚´ãƒªã‚ºãƒ ãŒã‚ã‚‹ã€‚å®Ÿè£…ã‚‚é›£ã—ãã¯ãªã•ãã†

algorithm

or

snow
ãƒªãƒ³ã‚¯
https://www.ibm.com/downloads/cas/DLPX0P1V
kiririmode 2023/04/24
3æ¬¡å…ƒç©ºé–“ã«ãŠã„ã¦ã‚³ãƒ³ãƒ†ãƒŠã®ç©ã¿æ–¹ã®æœ€é©åŒ–å•é¡Œã‚’ç¾å®Ÿæ™‚é–“ã§è§£ãã‚¢ãƒ«ã‚´ãƒªã‚ºãƒ ã®è«–æ–‡

or

algorithm
ãƒªãƒ³ã‚¯
ãƒ“ãƒ³ãƒ‘ãƒƒã‚ãƒ³ã‚°å•é¡Œã®è§£ãæ–¹ - Qiita
Deleted articles cannot be recovered. Draft of this article would be also deleted. Are you sure you want to delete this article?
kiririmode 2023/04/24
ç®±ã®ã‚µã‚¤ã‚ºãŒäº¤æ›å¯èƒ½ãªå ´åˆã¯ã‚¢ãƒ«ã‚´ãƒªã‚ºãƒ ã®åŠ¹çŽ‡ãŒè½ã¡ã‚‹ã€‚

or

algorithm
ãƒªãƒ³ã‚¯
ãƒ‘ãƒƒã‚ãƒ³ã‚°å•é¡Œ | opt100
kiririmode 2023/04/24
ç“¶ã®ã‚µã‚¤ã‚ºã‚’å¯å¤‰ã¨ã™ã‚‹å•é¡Œã¯ã€å¤‰å‹•ã‚µã‚¤ã‚ºãƒ™ã‚¯ãƒˆãƒ«ãƒ‘ãƒƒã‚ãƒ³ã‚°å•é¡Œ(variable size vector packing problem)ã¨è¨€ã‚ã‚Œã‚‹

or

algorithm
ãƒªãƒ³ã‚¯
ãƒ“ãƒ³ãƒ‘ãƒƒã‚ãƒ³ã‚°å•é¡Œ - Wikipedia
ãƒ“ãƒ³ãƒ‘ãƒƒã‚ãƒ³ã‚°å•é¡Œï¼ˆãƒ“ãƒ³ãƒ‘ãƒƒã‚ãƒ³ã‚°ã‚‚ã‚“ã ã„ï¼‰ã¨ã¯ã€é›¢æ•£æ•°å¦ã®çµ„åˆã›è«–ã®ä¸ã®NPå›°é›£å•é¡Œã§ã€ä¸Žãˆã‚‰ã‚ŒãŸã€Œè·ç‰©ï¼ˆé‡ã•ã‚„å€‹æ•°ãŒã¤ã„ã¦ã„ã‚‹ï¼‰ã€ã‚’ã¤ã‚ã‚‹ã€Œç®±ï¼ˆãƒ“ãƒ³ã‚„ã‚³ãƒ³ãƒ†ãƒŠãªã©ï¼‰ã€ã®æœ€å°æ•°ã‚’è¦‹ã¤ã‘ã‚‹ã‚‚ã®ã§ã‚ã‚‹ã€‚å•é¡Œã‚’è§£ããŸã‚ã«ãƒ“ãƒ³åž‹ï¼ˆç’çŠ¶åž‹ï¼‰ã®æ¨¡åž‹ã‚’ä½¿ã†ã®ã§ã“ã®ã‚ˆã†ã«å‘¼ã°ã‚Œã‚‹ã€‚ æ§˜ã€…ãªè§£æ±ºæ–¹æ³•ï¼ˆã‚¢ãƒ«ã‚´ãƒªã‚ºãƒ ï¼‰ãŒè€ƒæ¡ˆã•ã‚Œã¦ã„ã‚‹ãŒã€ã‚ã‚‰ã‚†ã‚‹å ´åˆã®ç®±ã®æœ€å°æ•°ã‚’åŠ¹çŽ‡çš„ã«è¦‹ã¤ã‘ã‚‹ã“ã¨ãŒã§ãã‚‹ã‚ˆã†ãªä¸‡èƒ½ãªã‚¢ãƒ«ã‚´ãƒªã‚ºãƒ ã¯ãªã„ï¼ˆNPå›°é›£å•é¡Œï¼‰ã€‚ 8å°ã®æ–°è»Šã‚’ãƒˆãƒ©ãƒƒã‚¯ã§ç§»å‹•ã™ã‚‹ã€‚æ–°è»Šã®é‡é‡ã¯ãã‚Œãžã‚Œ100ã‚ãƒã‚°ãƒ©ãƒ å˜ä½ã§ 33, 61, 58, 41, 50, 21, 60, 64 ã§ã‚ã‚‹ã€‚å„ãƒˆãƒ©ãƒƒã‚¯ãŒã€12,000 kg ã®é‡é‡ã¾ã§é‹ã¹ã‚‹ã¨ãã€å…¨ã¦ã®æ–°è»Šã‚’ä¸€åº¦ã«ç§»å‹•ã•ã›ã‚‹ã®ã«å¿…è¦ã¨ã•ã‚Œã‚‹ãƒˆãƒ©ãƒƒã‚¯ã®æœ€å°æ•°ã¯ã€ã„ãã¤ã§ã‚ã‚‹ã‹è€ƒãˆã‚‹ã€‚ã¾ãšã€ãƒˆãƒ©ãƒƒã‚¯ã‚’å®¹é‡120ã®ãƒ“ãƒ³ã¨ã—ã€æ–°è»Šã¯ã€ãã®ãƒ“ãƒ³ã«è©°ã‚ã‚‹è·ç‰©ã¨ã™ã‚‹ã€‚å…·ä½“çš„ã«èª¿ã¹ã‚‹
kiririmode 2023/04/24
ãƒ“ãƒ³ãƒ‘ãƒƒã‚ãƒ³ã‚°å•é¡Œã¯NPå›°é›£

or

algorithm
ãƒªãƒ³ã‚¯
A* - Wikipedia
A*æŽ¢ç´¢ã‚¢ãƒ«ã‚´ãƒªã‚ºãƒ A*ï¼ˆA-starã€ã‚¨ãƒ¼ã‚¹ã‚¿ãƒ¼ï¼‰æŽ¢ç´¢ã‚¢ãƒ«ã‚´ãƒªã‚ºãƒ ï¼ˆã‚¨ãƒ¼ã‚¹ã‚¿ãƒ¼ãŸã‚“ã•ãã‚¢ãƒ«ã‚´ãƒªã‚ºãƒ ï¼‰ã¯ã€ã‚°ãƒ©ãƒ•æŽ¢ç´¢ã‚¢ãƒ«ã‚´ãƒªã‚ºãƒ ã®ä¸€ã¤ã€‚ æœ€è‰¯å„ªå…ˆæŽ¢ç´¢ã‚’æ‹¡å¼µã—ãŸZ*ã«ã€ã•ã‚‰ã«få€¤ã¨ã—ã¦ã€Œç¾æ™‚ç‚¹ã¾ã§ã®è·é›¢ã€g ã¨ã€Œã‚´ãƒ¼ãƒ«ã¾ã§ã®æŽ¨å®šå€¤ã€h ã®å’Œã‚’æŽ¡ç”¨ã—ãŸã‚‚ã®[1]ã€‚h ã¯ ãƒ’ãƒ¥ãƒ¼ãƒªã‚¹ãƒ†ã‚£ãƒƒã‚¯é–¢æ•°ã¨å‘¼ã°ã‚Œã‚‹ã€‚ A* ã‚¢ãƒ«ã‚´ãƒªã‚ºãƒ ã¯ã€ã€Œã‚°ãƒ©ãƒ•ä¸Šã§ã‚¹ã‚¿ãƒ¼ãƒˆã‹ã‚‰ã‚´ãƒ¼ãƒ«ã¾ã§ã®é“ã‚’è¦‹ã¤ã‘ã‚‹ã€ã¨ã„ã†ã‚°ãƒ©ãƒ•æŽ¢ç´¢å•é¡Œã«ãŠã„ã¦ã€ ãƒ’ãƒ¥ãƒ¼ãƒªã‚¹ãƒ†ã‚£ãƒƒã‚¯é–¢æ•° h(n) ã¨ã„ã†æŽ¢ç´¢ã®é“æ¨™ã¨ãªã‚‹é–¢æ•°ã‚’ç”¨ã„ã¦æŽ¢ç´¢ã‚’è¡Œã†ã‚¢ãƒ«ã‚´ãƒªã‚ºãƒ ã§ã‚ã‚‹ã€‚h ã¯å„é ‚ç‚¹ n ã‹ã‚‰ã‚´ãƒ¼ãƒ«ã¾ã§ã®è·é›¢ã®ã‚ã‚‹å¦¥å½“ãªæŽ¨å®šå€¤ã‚’è¿”ã™é–¢æ•°ã§ã€è§£ãã‚°ãƒ©ãƒ•æŽ¢ç´¢å•é¡Œã®ç¨®é¡žã«å¿œã˜ã¦ã•ã¾ã–ã¾ãª h ã‚’è¨è¨ˆã™ã‚‹ã“ã¨ãŒå‡ºæ¥ã‚‹ã€‚ ä¾‹ãˆã°ã€ã‚«ãƒ¼ãƒŠãƒ“ãªã©ã§ç”¨ã„ã‚‰ã‚Œã‚‹å˜ç´”ãªäºŒæ¬¡å…ƒã®åœ°å›³ã§ã®æŽ¢ç´¢ã§ã¯ã€h ã¨ã—ã¦ãƒ¦ãƒ¼ã‚¯ãƒªãƒƒãƒ‰è·é›¢ ã‚’ä½¿ã†ã“ã¨ãŒã§ãã€ã“ã®å€¤ã¯é“ã«æ²¿ã£ãŸå®Ÿéš›ã®è·é›¢ã®ãŠãŠã¾
kiririmode 2023/04/16
A*

algorithm
ãƒªãƒ³ã‚¯
ã€Œé‡åã€ã¨çµ„åˆã›æœ€é©åŒ–ã«é–¢ã™ã‚‹æ€ªã—ã„è¨€èª¬ â€•ã¨ã‚ã‚‹ç ”ç©¶è€…ã®å°è¨€â€• - ã‚€ã—ã‚ƒãã—ã‚ƒã—ã¦ã‚„ã£ãŸï¼Œä»Šã¯åçœã—ã¦ã„ã‚‹æ—¥è¨˜
æœ€è¿‘ï¼Œé‡åã‚³ãƒ³ãƒ”ãƒ¥ãƒ¼ã‚¿ã®è©±é¡Œã‚’ãƒ‹ãƒ¥ãƒ¼ã‚¹ã‚„æ–°èžã§è¦‹ã‹ã‘ã‚‹ã“ã¨ãŒå¢—ãˆã¦ãã¾ã—ãŸï¼Ž ãã®ä¸ã§æ°—ã«ãªã£ã¦ããŸã®ãŒï¼Œçµ„åˆã›æœ€é©åŒ–ã¨é‡åã‚³ãƒ³ãƒ”ãƒ¥ãƒ¼ã‚¿ï¼ˆç‰¹ã«é‡åã‚¢ãƒ‹ãƒ¼ãƒªãƒ³ã‚°ï¼‰ã«é–¢ã™ã‚‹æ€ªã—ã„è¨€èª¬ï¼Žç§è‡ªèº«ã¯ï¼ˆå¤å…¸ã‚³ãƒ³ãƒ”ãƒ¥ãƒ¼ã‚¿ã§ã®ï¼‰çµ„åˆã›æœ€é©åŒ–ã®ç ”ç©¶ã‚’ã‚„ã£ã¦ãã¦ï¼Œé‡åã‚³ãƒ³ãƒ”ãƒ¥ãƒ¼ã‚¿ã‚’ç ”ç©¶ã—ã¦ã„ã‚‹ã‚ã‘ã§ã¯ãªã„ã®ã§ã™ãŒï¼Œã•ã™ãŒã«ã“ã‚Œã¯ã¡ã‚‡ã£ã¨ãƒ»ãƒ»ãƒ»ã¨æ€ã†è¨€èª¬ã‚’ä½•å›žã‹è¦‹ã‹ã‘ã¦ãã¾ã—ãŸï¼Ž æœ€è¿‘ã®ã€Œé‡åã€ã«å¯¾ã™ã‚‹éŽç†±ã¶ã‚Šã¯å‡„ã¾ã˜ã„ã®ã§ï¼Œã“ã†ã„ã†æ€ªã—ã„è¨€èª¬ãŒåºƒã¾ã‚‹ã®ã¯å›°ã‚Šã‚‚ã®ã§ã™ï¼Žã™ã§ã«Twitterä¸Šã«ã¯ï¼Œâ€œçµ„åˆã›æœ€é©åŒ–ã¯ä»Šã®ã‚³ãƒ³ãƒ”ãƒ¥ãƒ¼ã‚¿ã§ã¯è§£ã‘ãªã„â€ã¨ã‹â€œã§ã‚‚é‡åãªã‚‰ä¸€çž¬ã§è§£ã‘ã‚‹â€ã¨ã„ã†å‹˜é•ã„ã‚’ã—ã¦ã—ã¾ã£ã¦ã„ã‚‹äººãŒå¤šæ•°è¦‹ã‚‰ã‚Œã¾ã™*1ï¼Ž ã•ã™ãŒã«å±æ©Ÿæ„Ÿã‚’è¦šãˆã¦ããŸã®ã§ï¼Œã“ã®å ´ã§ãã¡ã‚“ã¨æŒ‡æ‘˜ã—ã¦ãŠãã“ã¨ã«ã—ã¾ã—ãŸï¼Ž ä»ŠåŒ—ç”£æ¥(TL;DR) â€œå¤å…¸ã‚³ãƒ³ãƒ”ãƒ¥ãƒ¼ã‚¿ã¯çµ„åˆã›æœ€é©åŒ–ã‚’è§£ã‘ãªã„â€ â†’ å¤å…¸ã‚³ãƒ³ãƒ”ãƒ¥ãƒ¼ã‚¿ã§çµ„åˆã›æœ€é©åŒ–ã‚’è§£
kiririmode 2021/08/14
é‡åã‚¢ãƒ‹ãƒ¼ãƒªãƒ³ã‚°ã¯åŽ³å¯†è§£ã‚’æ±‚ã‚ã‚‹ã‚ã‘ã§ã¯ãªããƒ’ãƒ¥ãƒ¼ãƒªã‚¹ãƒ†ã‚£ãƒƒã‚¯ã€æœ‰ç”¨ã ãŒç¾æ™‚ç‚¹ã§ã¯é©ç”¨ç¯„å›²ã¯é™ã‚‰ã‚Œã‚‹ã€‚TSPã§ã¯å¾“æ¥ã‚¢ãƒ«ã‚´ãƒªã‚ºãƒ ã®æ–¹ãŒé©ç”¨ç¯„å›²ãŒåºƒãã†

algorithm

quantum
ãƒªãƒ³ã‚¯
H.264ã®ç§˜å¯† | POSTD
(ç·¨æ³¨ï¼š2020/08/18ã€ã„ãŸã ã„ãŸãƒ•ã‚£ãƒ¼ãƒ‰ãƒãƒƒã‚¯ã‚’ã‚‚ã¨ã«è¨˜äº‹ã‚’ä¿®æ£ã„ãŸã—ã¾ã—ãŸã€‚) (2016/12/11ã€ã„ãŸã ãã¾ã—ãŸãƒ•ã‚£ãƒ¼ãƒ‰ãƒãƒƒã‚¯ã‚’ã‚‚ã¨ã«ç¿»è¨³ã‚’ä¿®æ£ã„ãŸã—ã¾ã—ãŸã€‚) H.264ã¯ã€å‹•ç”»åœ§ç¸®ã‚³ãƒ¼ãƒ‡ãƒƒã‚¯ã®æ¨™æº–è¦æ ¼ã§ã™ã€‚ãƒãƒƒãƒˆä¸Šã®å‹•ç”»ã€Blu-rayã€ã‚¹ãƒžãƒ›ã€ã‚»ã‚ãƒ¥ãƒªãƒ†ã‚£ã‚«ãƒ¡ãƒ©ã€ãƒ‰ãƒãƒ¼ãƒ³ãªã©ãªã©ã€ä»Šã‚„ã‚ã‚‰ã‚†ã‚‹ã¨ã“ã‚ã§H.264ãŒä½¿ã‚ã‚Œã¦ã„ã¾ã™ã€‚ H.264ã¯æ³¨ç›®ã™ã¹ãæŠ€è¡“ã®ã²ã¨ã¤ã§ã™ã€‚ãŸã£ãŸã²ã¨ã¤ã®ç›®æ¨™ã€ã¤ã¾ã‚Šãƒ•ãƒ«ãƒ¢ãƒ¼ã‚·ãƒ§ãƒ³ãƒ“ãƒ‡ã‚ªã®é€ä¿¡ã«è¦ã™ã‚‹ãƒãƒƒãƒˆãƒ¯ãƒ¼ã‚¯å¸¯åŸŸã‚’å‰Šæ¸›ã™ã‚‹ã“ã¨ã‚’ç›®æŒ‡ã—ãŸ30å¹´ä»¥ä¸Šã®åŠªåŠ›ã®çµæ™¶ãªã®ã§ã™ã€‚ æŠ€è¡“çš„ãªé¢ã§ã‚‚ã€H.264ã¯ã¨ã¦ã‚‚èˆˆå‘³æ·±ã„è¦æ ¼ã§ã™ã€‚ã“ã®è¨˜äº‹ã§ã¯ã€ãã®ä¸€éƒ¨ã«ã¤ã„ã¦æ¦‚è¦ãƒ¬ãƒ™ãƒ«ã§ã®çŸ¥è˜ã‚’å¾—ã‚‰ã‚Œã‚‹ã“ã¨ã§ã—ã‚‡ã†ã€‚ã‚ã¾ã‚Šè¤‡é›‘ã ã¨æ„Ÿã˜ã•ã›ãªã„ã‚ˆã†ã«ã™ã‚‹ã¤ã‚‚ã‚Šã§ã™ã€‚ä»Šå›žãŠã¯ãªã—ã™ã‚‹æ¦‚å¿µã®å¤šãã¯å‹•ç”»åœ§ç¸®å…¨èˆ¬ã«ã‚ã¦ã¯ã¾ã‚‹ã‚‚ã®ã§ã‚ã‚Šã€H.264ã«é™ã£ãŸã‚‚ã®ã§ã¯ã‚ã‚Šã¾ã›ã‚“
kiririmode 2017/10/08
H.264

algorithm
ãƒªãƒ³ã‚¯
ãƒ–ãƒãƒƒã‚¯ã‚¢ãƒ«ã‚´ãƒªã‚ºãƒ ã¨B-Treeã‚¢ãƒ«ã‚´ãƒªã‚ºãƒ
å¾Œã§èª¬æ˜Žã™ã‚‹ãŒã€ext2ã§ã¯ãƒ–ãƒãƒƒã‚¯ã‚°ãƒ«ãƒ¼ãƒ—0ã®ã€Œã‚¹ãƒ¼ãƒ‘ãƒ¼ãƒ–ãƒãƒƒã‚¯ã€ã«ã€ãƒ‘ãƒ¼ãƒ†ã‚£ã‚·ãƒ§ãƒ³å†…ã®ãƒ–ãƒãƒƒã‚¯ã‚°ãƒ«ãƒ¼ãƒ—å…¨ä½“ã®ç®¡ç†æƒ…å ±ãŒæ ¼ç´ã•ã‚Œã‚‹ã€‚ ã‚¹ãƒ¼ãƒ‘ãƒ¼ãƒ–ãƒãƒƒã‚¯ã«ã¯ã€ iãƒŽãƒ¼ãƒ‰ã®ç·æ•° ãƒ–ãƒãƒƒã‚¯ã‚µã‚¤ã‚º ãƒ•ã‚¡ã‚¤ãƒ«ã‚·ã‚¹ãƒ†ãƒ ã®ã‚µã‚¤ã‚º ç©ºããƒ–ãƒãƒƒã‚¯æ•°ã€ç©ºãiãƒŽãƒ¼ãƒ‰æ•° ä½¿ç”¨ã•ã‚Œã¦ã„ã‚‹ãƒ–ãƒãƒƒã‚¯ã‚°ãƒ«ãƒ¼ãƒ— æœ€çµ‚fsckæ™‚åˆ» ãªã©ã€ãƒ•ã‚¡ã‚¤ãƒ«ã‚·ã‚¹ãƒ†ãƒ ã«é–¢ã™ã‚‹ã•ã¾ã–ã¾ãªæƒ…å ±ãŒå«ã¾ã‚Œã¦ã„ã‚‹ã€‚ã€Œã‚°ãƒ«ãƒ¼ãƒ—ãƒ‡ã‚£ã‚¹ã‚¯ãƒªãƒ—ã‚¿ã€ã¯ã€ã€Œãƒ‡ãƒ¼ã‚¿ãƒ–ãƒãƒƒã‚¯ãƒ“ãƒƒãƒˆãƒžãƒƒãƒ—ã€ã‚„ã€ŒiãƒŽãƒ¼ãƒ‰ãƒ“ãƒƒãƒˆãƒžãƒƒãƒ—ã€æƒ…å ±ãªã©ã‚’å‚ç…§ã—ã¦ç©ºããƒ–ãƒãƒƒã‚¯ã‚„ç©ºãiãƒŽãƒ¼ãƒ‰ã‚’æŽ¢ã—ã€ãƒ‡ãƒ¼ã‚¿ã‚’å‰²ã‚Šå½“ã¦ã‚‹éš›ã«æœ€é©ãªãƒ–ãƒãƒƒã‚¯ã‚’æ±ºå®šã™ã‚‹ã®ã«ä½¿ã‚ã‚Œã‚‹ã€‚ ã€Œãƒ‡ãƒ¼ã‚¿ãƒ–ãƒãƒƒã‚¯ãƒ“ãƒƒãƒˆãƒžãƒƒãƒ—ã€ã¨ã€ŒiãƒŽãƒ¼ãƒ‰ãƒ“ãƒƒãƒˆãƒžãƒƒãƒ—ã€ã¯ã€ãƒ“ãƒƒãƒˆãƒžãƒƒãƒ—æƒ…å ±ã§ã‚ã‚‹ã€‚ãƒ“ãƒƒãƒˆãƒžãƒƒãƒ—æƒ…å ±ã¯ã€ã‚ã‚‹ãƒ‡ãƒ¼ã‚¿ãŒä½¿ç”¨ä¸ã‹æœªä½¿ç”¨ã‹ã‚’0ã‹1ã®bitã§ä¿æŒã—ã¦ã„ã‚‹ã€‚1byteï¼8bitsãªã®ã§ã€80å€‹ã®ãƒ–ãƒãƒƒã‚¯ã®çŠ¶æ…‹ã‚’10bytesã§ç®¡ç†ã§
kiririmode 2010/04/18
filesystem

algorithm

btree
ãƒªãƒ³ã‚¯
æ¨ªç€ãƒ—ãƒã‚°ãƒ©ãƒŸãƒ³ã‚° ç¬¬6å›ž: chatty: å°ã†ã‚‹ã•ã„ç«¯æœ«
æœ€çµ‚æ›´æ–°æ—¥: 2002-09-18 (å…¬é–‹æ—¥: 2002-09-18) Unix Magazine èªŒã« 2002å¹´1æœˆå·ã‹ã‚‰ 2003å¹´2æœˆå·ã«ã‹ã‘ã¦é€£è¼‰ã— ã¦ã„ãŸè¨˜äº‹ã®å…ƒã®åŽŸç¨¿ã§ã™ã€‚ æ‰æ°—ã«å¯Œã‚“ã ã“ã¨ã¯å€‹äººãŒè¡Œã†ã®ãŒé€šä¾‹ã§ã‚ã‚Šã€ä¿¡ã˜ãŒãŸããƒã‚« ã•åŠ æ¸›ã¯å¤§æŠµçµ„ç¹”ã«å¸°ã•ã‚Œã‚‹ã‚‚ã®ã§ã‚ã‚‹ã€‚ -- Jon Bentley *1 å½¹ã«ç«‹ãŸãªã„ã‚½ãƒ•ãƒˆã‚¦ã‚§ã‚¢ã‚’ä½œã‚‹ã®ãŒå¥½ãã ã€‚é¢å€’ãªä½œæ¥ã‚’æ¥½ã«ã™ ã‚‹æ¨ªç€ã‚½ãƒ•ãƒˆã‚¦ã‚§ã‚¢ã‚‚ã„ã„ãŒã€ãŸã¾ã«ã¯äººã‚’å‘†ã‚Œã•ã›ã‚‹ãã ã‚‰ãªã„ ã‚½ãƒ•ãƒˆã‚¦ã‚§ã‚¢ã‚’ä½œã‚‹ã®ã‚‚æ¥½ã—ã„ã€‚ ä»¥å‰ã«ç§ãŒé–‹ç™ºã—ãŸ cdbiff*2ã¨ã„ã†ã‚½ãƒ•ãƒˆ ã‚¦ã‚§ã‚¢ã¯ã€ãƒ¡ãƒ¼ãƒ«ãŒå±Šãã¨ PC ã® CD-ROMãƒ‰ãƒ©ã‚¤ãƒ–ãŒé–‹ã„ã¦ãƒ¡ãƒ¼ãƒ« ã®åˆ°ç€ã‚’é€šçŸ¥ã™ã‚‹ã¨ã„ã†å½¹ã«ç«‹ãŸãªã„ã‚‚ã®ã§ã‚ã£ãŸãŒã€ãã®ãã ã‚‰ ãªã•ãŒå—ã‘ã¦äºˆæƒ³å¤–ã®å¥½è©•ã‚’å¾—ãŸã€‚ä»Šå›žã¯ã€ãã†ã—ãŸå½¹ã«ç«‹ãŸãªã„ ã‚½ãƒ•ãƒˆã‚¦ã‚§ã‚¢ã® 1ã¤ã§ã‚ã‚‹ã€å°ã†ã‚‹ã•ã„ç«¯æœ« chatty*3 ã‚’ç´¹ä»‹ã™ã‚‹ã€‚
kiririmode 2008/10/05
TRIE

algorithm

*programming
ãƒªãƒ³ã‚¯
Spaghetti Source - å„ç¨®ã‚¢ãƒ«ã‚´ãƒªã‚ºãƒ ã® C++ ã«ã‚ˆã‚‹å®Ÿè£…
ACM/ICPCï¼ˆãƒ—ãƒã‚°ãƒ©ãƒŸãƒ³ã‚°ã‚³ãƒ³ãƒ†ã‚¹ãƒˆï¼‰ç³»åˆ—ã®å•é¡Œã‚’è§£ãã“ã¨ã‚’ç›®æ¨™ã«ã—ã¦ï¼Œå„ç¨®ã‚¢ãƒ«ã‚´ãƒªã‚ºãƒ ã‚’ C++ ã§å®Ÿè£…ã—ã¦ã¿ãŸï¼Žæ¥µã‚ã¦æ„åœ°ãŒæ‚ªã„é¡žã®å•é¡Œã«ã¯å¯¾å¿œã—ã¦ã„ãªã„ã—ï¼Œç‰¹å®šã®å…¥åŠ›ã«å¯¾ã—ã¦é«˜é€Ÿã«å‹•ãã¨ã„ã†ã“ã¨ã‚‚ãªã„ï¼Žè¨ˆç®—é‡ã‚‚æœ€è‰¯ã¨ã¯é™ã‚‰ãªã„ï¼Ž ã“ã‚Œã‚‰ã‚’å‚è€ƒã«ã™ã‚‹æ–¹ã¸ã®æ³¨æ„ã¨ãŠé¡˜ã„ï¼š ã“ã‚Œã‚‰ã®è¨˜è¿°ã¯æ£ç¢ºã¨ã¯é™ã‚Šã¾ã›ã‚“ï¼Žå‚è€ƒæ–‡çŒ®ã‚’å‚ç…§ã™ã‚‹ã“ã¨ã‚’å¼·ãæŽ¨å¥¨ã—ã¾ã™ï¼Žé–“é•ã£ã¦ã„ã‚‹å ´åˆã¯æ˜¯éžæ•™ãˆã¦ãã ã•ã„ï¼Ž ã“ã‚Œã‚‰ã®ãƒ—ãƒã‚°ãƒ©ãƒ ã¯é–“é•ã£ã¦ã„ã‚‹ã‹ã‚‚ã—ã‚Œã¾ã›ã‚“ï¼Žå„äººã§æ¤œè¨¼ã™ã‚‹ã“ã¨ã‚’å¼·ãæŽ¨å¥¨ã—ã¾ã™ï¼Žãƒã‚°ãŒã‚ã‚Œã°æ˜¯éžæ•™ãˆã¦ãã ã•ã„ï¼Ž åˆ†é¡žãŒæ€ªã—ã„ã®ã§ï¼Œã“ã‚Œã¯ã“ã£ã¡ã ã‚ã†ï¼Œã¨ã„ã†ã“ã¨ãŒã‚ã‚Œã°ã‚³ãƒ¡ãƒ³ãƒˆã‚’ä¸‹ã•ã‚‹ã¨åŠ©ã‹ã‚Šã¾ã™ï¼Ž æ³¨æ„ï¼ ç¾åœ¨æ›¸ãæ›ãˆä¸ TODO åˆ†é¡žã‚’æ£ã—ãè¡Œã†ï¼Ž å…¨ä½“çš„ã«èª¬æ˜Žã¨ä½¿ã„æ–¹ã‚’è©³ã—ãï¼Ž Verify ã—ã¦ã„ãªã„ã‚‚ã®ã‚’ Verifyï¼Ž ãƒœãƒãƒŽã‚¤å›³ï¼ˆã„ã¤ã«ãªã‚‹ã“ã¨ã‚„ã‚‰â€¦â€¦ï¼‰ åŸºæœ¬ ãƒ†ãƒ³ãƒ—ãƒ¬ãƒ¼ãƒˆ ã‚°ãƒ©ãƒ•
kiririmode 2008/08/15
*programming

c++

algorithm
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†
kiririmode 2008/04/05
ãªã‚“ã§ãƒ¡ãƒ¼ãƒ«ã®ãƒ•ã‚£ãƒ«ã‚¿ãƒªãƒ³ã‚°ã«ãƒ™ã‚¤ã‚ºãƒ•ã‚£ãƒ«ã‚¿ãŒä½¿ã‚ã‚Œã¦ã„ã‚‹ã‹ãŒã‚ˆãã‚ã‹ã‚‹

algorithm

spam
ãƒªãƒ³ã‚¯
1 2 æ¬¡ã®ãƒšãƒ¼ã‚¸