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

»çÀÌÆ®¸Ê

Loading..

Please wait....

±¹³» ÇÐȸÁö

Ȩ Ȩ > ¿¬±¸¹®Çå > ±¹³» ÇÐȸÁö > µ¥ÀÌÅͺ£À̽º ¿¬±¸È¸Áö(SIGDB)

µ¥ÀÌÅͺ£À̽º ¿¬±¸È¸Áö(SIGDB)

Current Result Document :

ÇѱÛÁ¦¸ñ(Korean Title) Àǹ̷ÐÀû ¹æ¹ý·ÐÀ» È°¿ëÇÑ À¥»çÀÌÆ® ¹æ¹®ÀÚÀÇ Àα¸Åë°èÇÐÀû Ư¼º ¿¹Ãø
¿µ¹®Á¦¸ñ(English Title) Prediction of Demographics of Web Site Audience Using Semantic Approach
ÀúÀÚ(Author) ±èÀÏÁÖ   Iljoo Kim  
¿ø¹®¼ö·Ïó(Citation) VOL 36 NO. 01 PP. 0039 ~ 0056 (2020. 04)
Çѱ۳»¿ë
(Korean Abstract)
À¥ÀÇ ²ÙÁØÈ÷ Áõ°¡ÇÏ´Â ±× Å©±â¿Í ´Ù¾ç¼ºÀº, À¥ ÄÜÅÙÃ÷ÀÇ °ø±ÞÀÚ, ¼ÒºñÀÚ, ±×¸®°í ±¤°íÁÖ ¸ðµÎ¿¡°Ô ¿©·¯ µµÀü°úÁ¦µéÀ» ´øÁø´Ù. ÀÌ·¯ÇÑ È¯°æ¿¡¼­, °³º° À¥ »çÀÌÆ®¿¡ ¹æ¹®ÇÏ´Â ¹æ¹®ÀÚµéÀÇ Àα¸Åë°èÇÐÀû Ư¼ºÀº ¸Å¿ì À¯¿ëÇÑ Á¤º¸ÀÌ´Ù. °³º° À¥»çÀÌÆ®ÀÇ Áֹ湮ÀÚµéÀÇ Àα¸Åë°èÇÐÀû Ư¼ºÀ» ¿¹ÃøÇÒ ¼ö ÀÖ´Ù¸é, ÀÌ´Â °ø±ÞÀÚ¿¡°Ô ÀÚ½ÅÀÇ ÄÜÅÙÃ÷°¡ ÀǵµÇÑ ¹æ¹®ÀÚµéÀ» È¿°úÀûÀ¸·Î ²ø¾îµéÀÏÁöÀÇ Á¡°ËÀ» °¡´ÉÄÉ ÇÒ °ÍÀ̸ç, ±¤°íÁÖ¿¡°Ô ±¤°íÈ¿°ú¸¦ ´õ¿í ³ô¿©ÁÙ À¥ ÆäÀÌÁö¸¦ ã°Ô ÇØ ÁÙ °ÍÀÌ´Ù. À̸¦ À§ÇØ, ±âÁ¸ÀÇ ¿¬±¸µéÀÌ ¿©·¯ ¹æ¹ý·ÐÀ» Á¦½ÃÇÏ¿´À¸³ª ´ëºÎºÐ ÄíÅ° ¹× »ç¿ëÀÚµî·ÏÀ» ÅëÇØ ¼öÁýµÈ Ŭ¸¯½ºÆ®¸² µ¥ÀÌÅÍ¿¡ ÀÇÁ¸ÇÏ¿© ±× È®À强 ¹× ºñ¿ë¿¡ ¹®Á¦°¡ ÀÖ°í, ±¤°íÁַμ­´Â ¾ò±â ¾î·Á¿î µ¥ÀÌÅͶó´Â ÇÑ°è°¡ ÀÖ´Ù. ÀÌ ¿¬±¸¿¡¼­´Â, ÇØ´ç À¥»çÀÌÆ®(ÆäÀÌÁö)·ÎºÎÅÍ Á÷Á¢ ±¸ÇÒ ¼ö ÀÖ´Â Áö¿ª½ÅÈ£(local cues), ƯÈ÷ À¥ÆäÀÌÁö¿¡¼­ Á÷Á¢ ÃßÃâ °¡´ÉÇÑ Àǹ̷ÐÀû ´Ü¼­µéÀ» È°¿ëÇÏ¿© Àα¸Åë°èÇÐÀû Ư¼ºÀ» ¿¹ÃøÇÏ´Â ¹æ¹ý·ÐÀ» Á¦¾ÈÇÏ°í È¿°ú¸¦ °ËÁõÇÏ°íÀÚ ÇÑ´Ù. ±× °á°ú, ±âÁ¸ ¹æ¹ý·ÐÀÇ ÇÑ°èÁ¡µéÀ» ±Øº¹ÇÑ ÀÌ ¹æ¹ý·ÐÀÌ ¿¹ÃøÀÇ Á¤È®¼º¿¡µµ È¿°úÀûÀÌ°í °æÀï·ÂÀÌ ÀÖÀ½À» ¾Ë ¼ö ÀÖ¾ú°í, ¶ÇÇÑ, ±âÁ¸ÀÇ ¹æ¹ý·Ð¿¡ ºñÇØ ±× È¿À²¼º¿¡ À־µµ ¿ì¼öÇÔÀ» È®ÀÎÇÏ¿´´Ù.
¿µ¹®³»¿ë
(English Abstract)
The rapidly changing and ever expanding nature of the Web poses great challenges for producers/consumers of content, and advertisers who want to place advertisements next to relevant content, and the demographics of the consumers who are likely to visit a given web site is a highly useful piece of information for the stakeholders. Previous works attempted to estimate the profile of likely consumers of web sites using the click stream/user registration information, but such data are costly to gather and evoke privacy/scalability concerns and not readily accessible to advertisers. Hence, in this research, we suggest an approach that overcomes the limitations by predicting the likely audience demographics of web sites based solely on the semantic information locally available on each web site. From the results, we could observe that, while overcoming the aforementioned limitations, the suggested approach achieves the highly competitive prediction performance as well as the significantly improved prediction efficiency compared to the existing approaches.
Å°¿öµå(Keyword) ±â°èÇнÀ   ¿¹ÃøºÐ¼®   ÅؽºÆ® ºÐ·ù   Semantic cues   À¥»çÀÌÆ® Àα¸Åë°èÇÐÀû Ư¼º   Àǹ̷ÐÀû ´Ü¼­   LDA   Machine learning   Predictive analytics   Text classification   Audience demographics  
ÆÄÀÏ÷ºÎ PDF ´Ù¿î·Îµå