• Àüü
  • ÀüÀÚ/Àü±â
  • Åë½Å
  • ÄÄÇ»ÅÍ
´Ý±â

»çÀÌÆ®¸Ê

Loading..

Please wait....

±¹³» ³í¹®Áö

Ȩ Ȩ > ¿¬±¸¹®Çå > ±¹³» ³í¹®Áö > Çѱ¹Á¤º¸°úÇÐȸ ³í¹®Áö > Á¤º¸°úÇÐȸ ÄÄÇ»ÆÃÀÇ ½ÇÁ¦ ³í¹®Áö (KIISE Transactions on Computing Practices)

Á¤º¸°úÇÐȸ ÄÄÇ»ÆÃÀÇ ½ÇÁ¦ ³í¹®Áö (KIISE Transactions on Computing Practices)

Current Result Document :

ÇѱÛÁ¦¸ñ(Korean Title) Áßø Á¶°ÇÀÌ ¾ø´Â ±ÔÄ¢ °Ë»ö ½Ã°£ ´ÜÃàÀ» À§ÇÑ ´Ù°èÃþ Àε¦½Ì ±â¼ú
¿µ¹®Á¦¸ñ(English Title) Multilevel Indexing for Reducing Searching Time of Rules without Overlapping Conditions
ÀúÀÚ(Author) ±è±¤¼ö   ¹Úµ¿È¯   Áø¼ºÀÏ   Kwangsoo Kim   Dong-Hwan Park   Seong-il Jin  
¿ø¹®¼ö·Ïó(Citation) VOL 27 NO. 02 PP. 0069 ~ 0075 (2021. 02)
Çѱ۳»¿ë
(Korean Abstract)
±ÔÄ¢ ±â¹Ý ½Ã½ºÅÛÀº ºñ±³Àû °£´ÜÇÑ ±ÔÄ¢À» »ç¿ëÇÏ¿© µ¥ÀÌÅͷκÎÅÍ À¯È¿ÇÑ Á¤º¸¸¦ ÃßÃâÇÏ´Â Àΰø Áö´É ÀÀ¿ë¿¡ »ç¿ëµÈ´Ù. ±ÔÄ¢ ±â¹Ý ½Ã½ºÅÛÀº ÀϹÝÀûÀ¸·Î ÇнÀÀÌ ¾øÀ¸¹Ç·Î ½Ã½ºÅÛ ¼³Ä¡ ÈÄ Áï½Ã ÀÛµ¿ÇÏ´Â ÀåÁ¡ÀÌ ÀÖÀ¸³ª, ±ÔÄ¢ ¸ÅĪ¿¡ º¹ÀâÇÑ Ã³¸®°¡ ÇÊ¿äÇÏ´Ù. ±âÁ¸ÀÇ ±ÔÄ¢ ¸ÅĪ ¹æ¹ýµéÀº ±ÔÄ¢µéÀÌ ¼­·Î ÁßøµÇ´Â Á¶°ÇÀ» °¡Áú ¶§ ºü¸£Áö¸¸, Á¶°ÇµéÀÌ ÁßøµÇÁö ¾Ê´Â °æ¿ì °Ë»ö ¼Óµµ°¡ ÀúÇϵȴÙ. º» ³í¹®¿¡¼­´Â ±ÔÄ¢ÀÇ Á¶°ÇÀÌ ÁßøµÇÁö ¾ÊÀ» ¶§ ±ÔÄ¢ °Ë»ö ½Ã°£À» ´ÜÃàÇϱâ À§ÇÑ ´Ù´Ü°è Àε¦½Ì ¹æ¹ýÀ» Á¦¾ÈÇÑ´Ù. Á¦¾ÈµÈ ¹æ¹ýÀº 2´Ü°è °Ë»ö Æ®¸®¸¦ »ç¿ëÇϸç, Æ®¸®ÀÇ °¢ ·¹º§¿¡¼­ ¼­·Î ´Ù¸¥ °ªÀ» »ç¿ëÇÏ¿© ±ÔÄ¢À» ºÐ·ùÇÑ´Ù. ù ¹ø° ´Ü°è¿¡¼­´Â Á¶°Ç¿¡ Æ÷ÇÔµÈ ¼¾½Ì °ªÀÇ ¼ö¸¦ »ç¿ëÇϸç, µÎ ¹ø° ´Ü°è¿¡¼­´Â °¢ ¼¾½Ì °ªÀÇ ¹üÀ§¸¦ »ç¿ëÇÏ¿´´Ù. ºÐ¼®Àû ¹æ¹ýÀ» ÅëÇØ ÃÖÀûÀÇ ¼­ºê ±×·ì ¼ö¸¦ Á¤ÀÇÇÏ¿´À¸¸ç, ½ÇÇè °á°ú¸¦ ÅëÇØ Á¦¾ÈµÈ ¹æ¹ýÀÇ ºü¸¥ °Ë»ö ½Ã°£À» È®ÀÎÇÏ¿´´Ù. °Ë»ö ½Ã°£Àº ±âÁ¸ÀÇ PHREAK ¹æ½Äº¸´Ù ¾à 3 ¹è Á¤µµ ºü¸£´Ù.
¿µ¹®³»¿ë
(English Abstract)
A rule-based system is used in artificial intelligence applications, which extract meaningful information from data with relatively simple rules. This system offers a benefit that it can be activated immediately after system installation as usually it does not require learning, but it needs complex processing in rule matching. Previous rule matching algorithms achieve high performance when the rules have many overlapping conditions, but if that is not the case, the searching speed slows down. This paper describes a multilevel indexing method for reducing the rule retrieval time when the conditions in rules do not overlap. The proposed method generates a two-level indexing tree using different values at each level to classify the rules. The number of sensing types in the conditions is used at the first step and the number of ranges of each sensing value is used at the second step. The optimal number of ranges is defined by an analytical method. The experimental results show that the proposed method can achieve a fast search time. Its searching time is about three times faster than that of the PHREAK method.
Å°¿öµå(Keyword) ±ÔÄ¢ °Ë»ö   °Ë»ö ¼Óµµ   ´Ù°èÃþ Æ®¸®   »óÀÌÇÑ ºÐ·ù ±âÁØ   rule searching   searching speed   multilevel tree   different classification values  
ÆÄÀÏ÷ºÎ PDF ´Ù¿î·Îµå