• 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¸°úÇÐȸ³í¹®Áö (Journal of KIISE)

Á¤º¸°úÇÐȸ³í¹®Áö (Journal of KIISE)

Current Result Document :

ÇѱÛÁ¦¸ñ(Korean Title) ½Å°æ¸Á ¹× ºñ½Å°æ¸Á ¿ÀÅäÀÎÄÚ´õ ±â¹Ý Ãßõ ¸ðµ¨ÀÇ ¼º´É ºñ±³ ¹× ºÐ¼®
¿µ¹®Á¦¸ñ(English Title) Performance Comparison and Analysis Between Neural and Non-neural Autoencoder-based Recommender Systems
ÀúÀÚ(Author) Á¤À±±â   ÀÌÁ¾¿í   Yoonki Jeong   Jongwuk Lee  
¿ø¹®¼ö·Ïó(Citation) VOL 47 NO. 11 PP. 1078 ~ 1085 (2020. 11)
Çѱ۳»¿ë
(Korean Abstract)
´Ù¾çÇÑ ºÐ¾ß¿¡ ½ÉÃþ ½Å°æ¸ÁÀÌ µµÀԵǾî ȹ±âÀûÀÎ ¼º´É °³¼±À» º¸ÀÌ°í ÀÖÀ¸³ª, ÃÖ±Ù ½ÉÃþ ½Å°æ ¸Á ±â¹Ý Ãßõ ¸ðµ¨ÀÇ ¼º´É °³¼±ÀÌ Å©°Ô º¸ÀÌÁö ¾Ê´Â´Ù´Â ÁÖÀåÀÌ ³ª¿À°í ÀÖ´Ù. ÀÌ¿Í °°Àº ¹®Á¦´Â Ãßõ ¿¬±¸¿¡ Åë¿ëµÇ´Â ½ÇÇè ȯ°æÀÇ ºÎÀç¿Í Á¦¾È ¸ðµ¨ ¼º´É¿¡ ´ëÇÑ ¾ö¹ÐÇÑ ºÐ¼® ºÎÀç¿¡ ±âÀÎÇÑ´Ù. º» ³í¹®¿¡¼­´Â 1) Ãßõ ¸ðµ¨ÀÇ °øÁ¤ÇÑ ºñ±³¸¦ À§ÇÑ ½ÇÇè ÇÁ·ÎÅäÄÝÀ» ±¸¼ºÇÏ°í, 2) Ãßõ ¸ðµ¨ÀÇ ÇÑ ÃàÀÎ ¿ÀÅäÀÎÄÚ´õ ±â¹Ý Ãßõ ¸ðµ¨¿¡ ´ëÇؼ­ ½ÇÇèÀû °ËÁõÀ» ¼öÇàÇϸç, 3) »ç¿ëÀÚ¿Í Ç׸ñ Àα⵵¸¦ ±âÁØÀ¸·Î ¿©·¯ °³ÀÇ ¼¼ºÎ ±×·ìÀ¸·Î ³ª´©¾î ½ÇÇè °á°ú¸¦ ºÐ¼®ÇÑ´Ù. ½ÇÇè °á°ú, ¸ðµç µ¥ÀÌÅͼ¿¡¼­ ½Å°æ¸Á ±â¹Ý ¸ðµ¨ÀÇ Ãßõ ¼º´ÉÀÌ ºñ½Å°æ¸Á ´ëºñ ÀÏ°üÀûÀÎ ¼º´É °³¼±À» º¸ÀÌÁö ¾Ê¾ÒÀ¸¸ç, ½Å°æ¸Á ¸ðµ¨ ³»¿¡¼­µµ ÁÖµÈ Á¤È®µµ °³¼±À» È®ÀÎÇÒ ¼ö ¾ø¾ú´Ù. ÇÑÆí, ¼¼ºÎ ±×·ìº° ¼º´É Æò°¡ °á°ú¿¡¼­´Â Àαâ Ç׸ñ¿¡¼± ºñ½Å°æ¸Á ¸ðµ¨ÀÇ, ºñÀαâ Ç׸ñ¿¡¼± ½Å°æ¸Á ¸ðµ¨ÀÇ Á¤È®µµ°¡ ³ôÀ½ÀÌ È®ÀÎÇÏ¿´°í, À̸¦ ÅëÇØ ½Å°æ¸Á ¸ðµ¨ÀÇ º¹À⼺ÀÌ ºñÀαâ Ç׸ñ¿¡ ´ëÇÑ Ãßõ¿¡ µµ¿òÀÌ µÉ ¼ö ÀÖ´Ù°í ÆǴܵȴÙ.
¿µ¹®³»¿ë
(English Abstract)
While deep neural networks have been bringing advances in many domains, recent studies have shown that the performance gain from deep neural networks is not as extensive as reported, compared to the higher computational complexity they require. This phenomenon is caused by the lack of shared experimental settings and strict analysis of proposed methods. In this paper, 1) we build experimental settings for fair comparison between the different recommenders, 2) provide empirical studies on the performance of the autoencoder-based recommender, which is one of the main families in the literature, and 3) analyze the performance of a model according to user and item popularity. With extensive experiments, we found that there was no consistent improvement between the neural and the non-neural models in every dataset and there is no evidence that the non-neural models have been improving over time. Also, the non-neural models achieved better performance on popular item accuracy, while the neural models relatively perform better on less popular items.
Å°¿öµå(Keyword) Ãßõ ½Ã½ºÅÛ   ½ÉÃþ ½Å°æ¸Á   ¿ÀÅäÀÎÄÚ´õ   ¼º´É Æò°¡   Àα⵵   recommender system   deep neural networks   autoencoder   performance evaluation   popularity  
ÆÄÀÏ÷ºÎ PDF ´Ù¿î·Îµå