• 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 ³í¹®Áö C : ÄÄÇ»ÆÃÀÇ ½ÇÁ¦

Á¤º¸°úÇÐȸ ³í¹®Áö C : ÄÄÇ»ÆÃÀÇ ½ÇÁ¦

Current Result Document : 13 / 13

ÇѱÛÁ¦¸ñ(Korean Title) Æ®À§ÅÍ ¹®¼­ ºÐ¼®À» ÅëÇÑ °¨Á¤ ±â¹ÝÀÇ À½¾Ç Ãßõ ½Ã½ºÅÛ
¿µ¹®Á¦¸ñ(English Title) Emotion-based Music Recommendation System based on Twitter Document Analysis
ÀúÀÚ(Author) ÃÖÈ«±¸   ȲÀÎÁØ   Hong-gu Choi   Eenjun Hwang  
¿ø¹®¼ö·Ïó(Citation) VOL 18 NO. 11 PP. 0762 ~ 0767 (2012. 11)
Çѱ۳»¿ë
(Korean Abstract)
´Ù¾çÇÏ°í ¹æ´ëÇÑ ¾çÀÇ ¸ÖƼ¹Ìµð¾î ÄÜÅÙÃ÷°¡ º¸ÆíÈ­µÇ¸é¼­ À̸¦ È¿°úÀûÀ¸·Î È°¿ëÇϱâ À§ÇÑ ´Ù¾çÇÑ ¿¬±¸°¡ ¼öÇàµÇ°í ÀÖ´Ù. ƯÈ÷ À½¾ÇÀÇ °æ¿ì »ç¶÷ÀÇ °¨Á¤À̳ª ±âºÐ¿¡ µû¶ó µè°í ½ÍÀº À½¾ÇÀÇ À¯ÇüÀÌ Å©°Ô ´Þ¶óÁú ¼ö Àֱ⠶§¹®¿¡ À½¾Ç Ãßõ°ú °°Àº À½¾Ç °ü·Ã ¼­ºñ½º¸¦ À§Çؼ­´Â »ç¶÷ÀÇ °¨Á¤ ÆľÇÀÌ Áß¿äÇÏ´Ù. ÇÑÆí, ¼Ò¼È ³×Æ®¿öÅ© ¼­ºñ½º°¡ È®»êµÇ¸é¼­ ÀÏ»ó »ýÈ°À̳ª Á¤Ä¡Àû »ç°Ç, ¿µÈ­, Á¦Ç° µî ´Ù¾çÇÑ À̽´¿¡ ´ëÇÑ °³ÀÎÀÇ »ý°¢À̳ª ´À³¦À» Ç¥ÇöÇÏ°í °øÀ¯ÇÏ´Â °ÍÀÌ ºó¹øÇØÁ³´Ù. ƯÈ÷, ¸¶ÀÌÅ©·Î ºí·Î±×¿Í °°Àº °æ¿ì »ç¿ëÀÚÀÇ °¨Á¤À» ¾Ï½ÃÇÏ´Â ³»¿ëÀÌ Ç³ºÎÇϹǷÎ, ÀÌ·¯ÇÑ µ¥ÀÌÅ͸¦ ¼öÁýÇÏ°í ºÐ¼®ÇÏ¸é °³ÀÎÀÇ °ßÇسª °¨Á¤À» ÆľÇÇÏ´Â °ÍÀÌ °¡´ÉÇÏ´Ù. ÀÌ·¯ÇÑ °¨Á¤ ÃßÃâ°ú À½¾Ç ÃßõÀ» ¿¬°áÇÒ ¼ö ÀÖ´Ù¸é ´Ù¾çÇÑ ÇüÅÂÀÇ °¨Á¤ ±â¹Ý »ç¿ëÀÚ ¼­ºñ½º¸¦ È¿°úÀûÀ¸·Î Áö¿øÇÒ ¼ö ÀÖ´Ù. º» ³í¹®¿¡¼­´Â ´ëÇ¥ÀûÀÎ SNSÀÎ Æ®À§ÅÍ¿¡¼­ »ç¿ëÀÚÀÇ °¨Á¤À» ºÐ¼®ÇÏ°í ºÐ¼®ÇÑ °¨Á¤¿¡ ÀûÇÕÇÑ À½¾ÇÀ» ÃßõÇÏ´Â ±â¹ýÀ» Á¦¾ÈÇÏ´Ù. ±×¸®°í ½ÇÇèÀ» ÅëÇØ Á¦¾ÈÇÑ ±â¹ýÀÇ È¿À²¼ºÀ» °ËÁõÇÑ´Ù.
¿µ¹®³»¿ë
(English Abstract)
As a vast amount of diverse multimedia contents have been available, many works have been done to utilize them effectively. Especially, in the case of music, since the music type to which a user wants to listen depends on his emotion or mood, it has become very important to recognize user emotion for music-related services such as music recommendation. On the other hand, with the wide spread of social network services, it has become quite common for people to express and share personal thoughts or feelings on the various issues such as daily life, political events, movies or commercial products. In particular, as micro-blogs contain various emotion-rich resources, personal opinion or emotion can be extracted and recognized by collecting and analyzing such data. In order to support various emotion-based user services, mood recognition and music recommendation should be connected to each other. In this paper, we first show how to extract user emotion from twitter documents and then propose how to recommend music based on the extracted user emotion. Through experiments, we show that our scheme can produce satisfactory result.
Å°¿öµå(Keyword) Æ®À§ÅÍ   À½¾Ç Ãßõ   °¨Á¤ÃßÃâ   SNS   Twitter   Music Recommendation   Emotion Extraction  
ÆÄÀÏ÷ºÎ PDF ´Ù¿î·Îµå