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

»çÀÌÆ®¸Ê

Loading..

Please wait....

±¹³» ³í¹®Áö

Ȩ Ȩ > ¿¬±¸¹®Çå > ±¹³» ³í¹®Áö > Çѱ¹ÀÎÅͳÝÁ¤º¸ÇÐȸ ³í¹®Áö

Çѱ¹ÀÎÅͳÝÁ¤º¸ÇÐȸ ³í¹®Áö

Current Result Document : 7 / 38 ÀÌÀü°Ç ÀÌÀü°Ç   ´ÙÀ½°Ç ´ÙÀ½°Ç

ÇѱÛÁ¦¸ñ(Korean Title) º£ÀÌÁö¾È È®·ü ¹× Æó¼â ¼øÂ÷ÆÐÅÏ ¸¶ÀÌ´× ¹æ½ÄÀ» ÀÌ¿ëÇÑ ¼³¸í°¡´ÉÇÑ ·Î±× ÀÌ»óŽÁö ½Ã½ºÅÛ
¿µ¹®Á¦¸ñ(English Title) An Interpretable Log Anomaly System Using Bayesian Probability and Closed Sequence Pattern Mining
ÀúÀÚ(Author) À±Áö¿µ   ½Å°ÇÀ±   ±èµ¿¿í   ±è»ó¼ö   ÇÑ¸í¹¬   Jiyoung Yun   Gun-Yoon Shin   Dong-Wook Kim   Sang-Soo Kim   Myung-Mook Han  
¿ø¹®¼ö·Ïó(Citation) VOL 22 NO. 02 PP. 0077 ~ 0087 (2021. 04)
Çѱ۳»¿ë
(Korean Abstract)
ÀÎÅͳݰú °³Àοë ÄÄÇ»ÅÍ°¡ ¹ß´ÞÇϸ鼭 ´Ù¾çÇÏ°í º¹ÀâÇÑ °ø°ÝµéÀÌ µîÀåÇϱ⠽ÃÀÛÇß´Ù. °ø°ÝµéÀÌ º¹ÀâÇØÁü¿¡ µû¶ó ±âÁ¸¿¡ »ç¿ëÇÏ´ø ½Ã±×´Ïó ±â¹ÝÀÇ Å½Áö ¹æ½ÄÀ¸·Î ŽÁö°¡ ¾î·Á¿öÁ³À¸¸ç À̸¦ ÇØ°áÇϱâ À§ÇØ ÇàÀ§±â¹ÝÀÇ Å½Áö¸¦ À§ÇÑ ·Î±× ÀÌ»óŽÁö¿¡ ´ëÇÑ ¿¬±¸°¡ ÁÖ¸ñ ¹Þ±â ½ÃÀÛÇß´Ù. ÃÖ±Ù ·Î±× ÀÌ»óŽÁö¿¡ ´ëÇÑ ¿¬±¸´Â µö·¯´×À» È°¿ëÇØ ¼ø¼­¸¦ ÇнÀÇÏ´Â ¹æ½ÄÀ¸·Î ÀÌ·ç¾îÁö°í ÀÖÀ¸¸ç ÁÁÀº ¼º´ÉÀ» º¸¿©ÁØ´Ù. ÇÏÁö¸¸ ÁÁÀº ¼º´É¿¡µµ ºÒ±¸ÇÏ°í ÆÇ´Ü¿¡ ´ëÇÑ ±Ù°Å¸¦ Á¦°øÇÏÁö ¸øÇÑ´Ù´Â ÇÑ°èÁ¡À» Áö´Ñ´Ù. ÆÇ´Ü¿¡ ´ëÇÑ ±Ù°Å ¹× ¼³¸íÀ» Á¦°øÇÏÁö ¸øÇÒ °æ¿ì, µ¥ÀÌÅÍ°¡ ¿À¿°µÇ°Å³ª ¸ðµ¨ ÀÚü¿¡ °áÇÔÀÌ ¹ß»ýÇصµ À̸¦ ¹ß°ßÇϱ⠾î·Æ´Ù´Â ¹®Á¦Á¡À» Áö´Ñ´Ù. °á·ÐÀûÀ¸·Î »ç¿ëÀÚÀÇ ½Å·Ú¼ºÀ» ÀÒ°Ô µÈ´Ù. À̸¦ ÇØ°áÇϱâ À§ÇØ º» ¿¬±¸¿¡¼­´Â ¼³¸í°¡´ÉÇÑ ·Î±× ÀÌ»óŽÁö ½Ã½ºÅÛÀ» Á¦¾ÈÇÑ´Ù. º» ¿¬±¸´Â °¡Àå ¸ÕÀú ·Î±× ÆĽÌÀ» ÁøÇàÇØ ·Î±× Àü󸮸¦ ¼öÇàÇÑ´Ù. ÀÌÈÄ Àüó¸®µÈ ·Î±×µéÀ» ÀÌ¿ëÇØ º£ÀÌÁö¾È È®·ü ±â¹Ý ¼øÂ÷ ±ÔÄ¢ÃßÃâÀ» ÁøÇàÇÑ´Ù. °á°úÀûÀ¸·Î ¡°If Á¶°Ç then °á°ú, »çÈÄÈ®·ü(¥è)¡± Çü½ÄÀÇ ±ÔÄ¢ÁýÇÕÀ» ÃßÃâÇϸç ÀÌ¿Í ¸ÅĪµÉ °æ¿ì Á¤»ó, ¸ÅĪµÇÁö ¾ÊÀ» °æ¿ì, ÀÌ»óÇàÀ§·Î ÆÇ´ÜÇÏ°Ô µÈ´Ù. ½ÇÇèÀ¸·Î´Â HDFS ·Î±× µ¥ÀÌÅͼÂÀ» È°¿ëÇßÀ¸¸ç, ±× °á°ú F1score 92.7%ÀÇ ¼º´ÉÀ» ³ªÅ¸³»¾ú´Ù.
¿µ¹®³»¿ë
(English Abstract)
With the development of the Internet and personal computers, various and complex attacks begin to emerge. As the attacks become more complex, signature-based detection become difficult. It leads to the research on behavior-based log anomaly detection. Recent work utilizes deep learning to learn the order and it shows good performance. Despite its good performance, it does not provide any explanation for prediction. The lack of explanation can occur difficulty of finding contamination of data or the vulnerability of the model itself. As a result, the users lose their reliability of the model. To address this problem, this work proposes an explainable log anomaly detection system. In this study, log parsing is the first to proceed. Afterward, sequential rules are extracted by Bayesian posterior probability. As a result, the "If condition then results, post-probability" type rule set is extracted. If the sample is matched to the ruleset, it is normal, otherwise, it is an anomaly. We utilize HDFS datasets for the experiment, resulting in F1score 92.7% in test dataset.
Å°¿öµå(Keyword) ¼³¸í°¡´ÉÇÑ ÀΰøÁö´É   ·Î±× ÀÌ»óŽÁö ½Ã½ºÅÛ   º£ÀÌÁö¾È È®·ü   ±ÔÄ¢ ÃßÃâ   Explainable AI   Log anomaly detection   Bayesian probability   Rule extraction  
ÆÄÀÏ÷ºÎ PDF ´Ù¿î·Îµå