Andrei Andreyevich Markov

 (1856 ~ 1922)

·¯½Ã¾Æ ¼öÇÐÀÚ. ·ªÀÜ Ãâ»ý. 1878³â »óÆ®ÆäÅ׸£ºÎ¸£Å©´ëÇÐÀ» Á¹¾÷ÇÏ°í, °°Àº ÇØ ¸ð±³ÀÇ °­»ç, 96³â °úÇоÆÄ«µ¥¹Ì ȸ¿ø, 98³â ¸ð±³ÀÇ ±³¼ö°¡ µÇ¾ú´Ù. ¸¶¸£ÄÚÇÁ´Â £¼¾î¶² °èÅëÀÇ ½Ã°£Àû ¹ßÀüÀÇ ¸ð¾çÀÌ °èÅëÀÇ ÇöÀç »óÅÂ¿Í ¸ñÇ¥ÇÏ´Â ¹Ì·¡ÀÇ ½ÃÁ¡¸¸À¸·Î È®·ü·ÐÀûÀ¸·Î °áÁ¤µÇ´Â È®·ü°úÁ¤£¾À» µµÀÔÇÏ¿´´Ù. ÀÌ°ÍÀº ¿À´Ã³¯ £¼¸¶¸£ÄÚÇÁ°úÁ¤ : Markov process£¾À̶ó ºÒ¸®°í ÀÖ´Ù. Àú¼­¿¡ ¡¶Á¤Â÷¹ý(ïÒó¬Ûö, 1896)¡·°ú ¡¶È®·ü·Ð(1912)¡·ÀÌ ÀÖ´Ù.

Markov Model : ¼­¿ï´ë Àü»ê¾ð¾îÇÐ ¿¬±¸½Ç : ÀÌ À¯ÇÑ»óÅ ¿ÀÅ丶Ÿ¿¡ ´Ù½Ã À§¿Í °°ÀÌ È®·üÀ» ºÙÀÎ °ÍÀ» "È®·üÀû À¯ÇÑ»óÅ ¿ÀÅ丶Ÿ(probabilistic finite state automaton)À̶ó°í Çϸç, Åë°èÇп¡¼­´Â À̸¦ Markov ChainÀ̶ó°í ÇÑ´Ù. ÀÌ´Â À¯ÇÑ»óÅ ¿ÀÅ丶Ÿ¿Í °°À¸³ª °¢°¢ÀÇ ¿ø (»óÅÂ)ÀÌ ÃëÇØÁö´Â °ÍÀº È®·ü¿¡ ÀÇÇÑ´Ù´Â Á¡¿¡ Â÷ÀÌ°¡ ÀÖ´Ù. µû¶ó¼­ °¢°¢ÀÇ »óÅ¿¡¼­ ´ÙÀ½ »óÅ·Π°¡°ÔµÇ´Â arcÀÇ È®·üÀÇ ÇÕÀº  ¹Ýµå½Ã 1ÀÌ µÇ¾î¾ß ÇÑ´Ù. ÀÌ·± Markov ¸ðµ¨Àº À¯ÇÑ»óÅÂÀÇ ¿ÀÅ丶Ÿ¿¡ ±âÃÊÇϹǷÎ, ´ÙÀ½ »óÅ°¡ ¸¶Áö¸· »óÅ°¡ µÇ´Â °ÍÀº ÇöÀç ¾î´À »óÅ¿¡ ÀÖ´À³Ä¿¡ µû¶ó °áÁ¤µÈ´Ù. ÀÌ Markov ModelÀº ´Ù½Ã, ¾î¶² »óŸ¦ Áö³ª°¡´ÂÁö ¾Ë ¼ö ÀÖ°í µû¶ó¼­ »óÅÂÀÇ ¼ø¼­³ª ¾î¶² °áÁ¤ ÇÔ¼ö (deterministic function)¸¦ ¾Ë ¼ö ÀÖ´Â Visible Markov Model °ú  Áö³ª°¡´Â »óŸ¦ ¾ËÁö ¸øÇÏ°í ´ÜÁö ¾î¶² È®·üÀûÀÎ function¸¸À» ¾Ë ¼ö ÀÖ´Â Hidden Markov Model·Î ±¸ºÐµÉ ¼ö ÀÖ´Ù. ....

¸¶¸£ÄÚÇÁ ¾Ë°í¸®Áò (Markov Algorithm)   Àº´Ð ¸¶¸£ÄÚÇÁ ¸ðµ¨ (Hidden Markov Model)   À½¼ºÀÎ½Ä (Speech Recognition)   ÀÚ¿¬¾îó¸® (Natural Language Processing)   ºÒÈ®½Ç¼º (Uncertainty)   È®·ü (Probability)

Andrei Markop : ¸¶¸£ÄÚÇÁ ¸ðµ¨, ¸¶¸£ÄÚÇÁ üÀÎ

"Markov is particularly remembered for his study of Markov chains, sequences of random variables in which the future variable is determined by the present variable but is independent of the way in which the present state arose from its predecessors."...... University of St Andrews, Scotland

"His name is best known for the concept of the Markov chain, a series of events in which the probability of a given event occurring depends only on the immediately previous event."

"A related technique, called Hidden Markov models, allows probability to anticipate sequences. A speech recognition application, for example, knows that the sound most likely to follow "q" is "u." Along those lines, the software can also calculate the possible utterance of the word Qagga, an extinct zebra."

Hidden Markov Models Tutorial from the School of Computing, University of Leeds. "[T]he type of system we will consider in this tutorial. * First we will introduce systems which generate probabalistic patterns in time, such as the weather fluctuating between sunny and rainy. * We then look at systems where what we wish to predict is not what we observe - the underlying system is hidden. In the above example, the observed sequence would be the seaweed and the hidden system would be the actual weather. * We then look at some problems that can be solved once the system has been modeled."