• 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) ¿µ»ó ÄÜÅÙÃ÷ ¿ÀºêÁ§Æ®ÀÇ °¨Á¤ ¿µÇâ·Â ºÐ¼®À» ÅëÇÑ ¿ÀºêÁ§Æ® °¨Á¤ ¼øÀ§ °áÁ¤¹ý
¿µ¹®Á¦¸ñ(English Title) Objects Sentiment Ranking Method Using Sentimental Sensitivity Analysis of Objects in Video Content
ÀúÀÚ(Author) À̼ÒÁ¤   ¼Õ±ÔÁø   ±èÀ±Èñ   Sojeong Lee   Gyujin Son   Yoonhee Kim                          
¿ø¹®¼ö·Ïó(Citation) VOL 26 NO. 12 PP. 0528 ~ 0534 (2020. 12)
Çѱ۳»¿ë
(Korean Abstract)
Çö´ë¿¡´Â ´Ù¾çÇÑ ±â¼úÀÇ ¹ßÀüÀ¸·Î µ¿¿µ»ó ÄÜÅÙÃ÷°¡ Áõ°¡ÇÏ°í ÀÖ°í, ÀÌ¿¡ µû¶ó »ç¶÷µéÀº ÀÚ½ÅÀÇ °¨Á¤ Ç¥Çö¿¡ ÀÖ¾î µ¿¿µ»ó ¸Åü¸¦ È°¿ëÇÏ´Â °ÍÀÌ ¸¸¿¬ÇØÁ³´Ù. µû¶ó¼­ ÅؽºÆ® ºÐ¼®»Ó¸¸ ¾Æ´Ï¶ó ¹ÝÀÀ µ¿¿µ»ó ¼Ó ½ÃûÀÚÀÇ °¨Á¤À» ºÐ¼®ÇÒ ÇÊ¿ä°¡ ÀÖ´Ù. ±âÁ¸ÀÇ ½Ã½ºÅÛÀº ÄÜÅÙÃ÷ÀÇ °ü·Ã ¹ÝÀÀ µ¿¿µ»ó ¼Ó ½ÃûÀÚÀÇ °¨Á¤°ú Á÷°áÇÏ¿© ¿ÀºêÁ§Æ®ÀÇ °¨Á¤°ªÀ» µµÃâÇÏÁö´Â ¾ÊÀ¸¹Ç·Î ¿ÀºêÁ§Æ®ÀÇ °¨Á¤°ªÀ» ÃßÃâÇÒ ¹æ¹ýÀÌ ÇÊ¿äÇÏ´Ù. µû¶ó¼­ º» ¿¬±¸¿¡¼­´Â ÄÜÅÙÃ÷ÀÇ µîÀå ¿ÀºêÁ§Æ®¿Í °ü·Ã ¹ÝÀÀ µ¿¿µ»ó ¿ÀºêÁ§Æ®ÀÇ °¨Á¤ ¿µÇâ·ÂÀ» ºÐ¼®ÇÏ°í ¿ÀºêÁ§Æ® °¨Á¤ ¼øÀ§¸¦ °áÁ¤ÇÒ ¼ö ÀÖ´Â ±â¹ýÀ» Á¦¾ÈÇÑ´Ù. º» ½Ã½ºÅÛÀº °¨Á¤ ¿ÀºêÁ§Æ® ÃßÃâÀ» À§ÇØ ´ë»ó ÄÜÅÙÃ÷¿¡ ´ëÇÑ ¹ÝÀÀ µ¿¿µ»ó ¼öÁýÀ» ÅëÇØ ½ÃûÀÚ °¨Á¤À» ºÐ¼®ÇÏ°í, ´ë»ó ÄÜÅÙÃ÷ ¼Ó ¿ÀºêÁ§Æ®ÀÇ ½ÃûÀÚ °¨Á¤°ªÀ» Á¤·®È­ÇÑ´Ù. ¶ÇÇÑ, ÃßÃâµÈ ¿ÀºêÁ§Æ®µéÀ» ±â¹ÝÀ¸·Î °¨Á¤º° ¿ÀºêÁ§Æ® »çÀüÀ» ±¸ÃàÇÑ´Ù. ´Ù¾çÇÑ µ¿¿µ»óÀ» ´ë»óÀ¸·Î ÇÑ ½ÇÇèÀ» ÅëÇØ ±¸°£ °¡ÁßÄ¡¸¦ Àû¿ëÇÑ ±¸°£ÀÇ ¿ÀºêÁ§Æ® Á¸Àç º¯È­¿¡ µû¶ó ´ÙÁß ¿ÀºêÁ§Æ® °¡ÁßÄ¡¸¦ Àû¿ëÇÏ¿© °¨Á¤ ¿µÇâ·Â Æò°¡¸¦ ÁøÇàÇÔÀ¸·Î½á ¼±º°Àû °¨Á¤ ¿ÀºêÁ§Æ® ÃßÃâÀÌ °¡´ÉÇÔÀ» º¸¿´´Ù.
¿µ¹®³»¿ë
(English Abstract)
Currently, video content is in demand because of the development of various technologies, that have led to the widespread use of video media for people to express their thoughts. Thus, it is necessary to analyze the viewer's thoughts in the reaction video as well as text analysis. The existing system does not derive the sentimental value of the objects directly from the viewer's thoughts in the related reaction video of the content, so an approach to extracting the objects sentiment is needed. Thus, this study analyzed the sentimental influence of the content's appearance objects and associated reaction video objects and proposes techniques to extract the ranking of the objects sentiment. This system analyzed viewers' sentiment through collecting reaction videos to target content for extracting sentiment objects and quantifies viewer sentiment values of objects in the target content. A sentimental object dictionary was built based on the extracted objects. The experiments showed to collect effective sentimental objects selectively, applying multi-level weights to the analysis of sentiment objects depending on the change of their existence in specific periods of videos.
Å°¿öµå(Keyword) À̹ÌÁö µö·¯´×   °¨Á¤ ¿µÇâ·Â ºÐ¼®   °¨Á¤ ¿ÀºêÁ§Æ®   À¯Æ©ºê   µ¿¿µ»ó   image deep learning   sentimental sensitivity analysis   sentimental object   YouTube   video                    
ÆÄÀÏ÷ºÎ PDF ´Ù¿î·Îµå