µ¥ÀÌÅͺ£À̽º ¿¬±¸È¸Áö(SIGDB)
Current Result Document :
ÇѱÛÁ¦¸ñ(Korean Title) |
½Å±Ô ½º¸¶Æ®Æù »ç¿ëÀÚÀÇ Áß´Ü°¡´É¼º ¿¹Ãø Á¤È®µµ Çâ»ó ¿¬±¸ |
¿µ¹®Á¦¸ñ(English Title) |
Study on Improving Interruptibility Prediction Accuracy for New Smartphone Users |
ÀúÀÚ(Author) |
¾çÇö¼®
Ãֹμö
ÀÌÀç±æ
HyunSeok Yang
Minsu Choy
Jae-Gil Lee
|
¿ø¹®¼ö·Ïó(Citation) |
VOL 32 NO. 02 PP. 0119 ~ 0129 (2016. 08) |
Çѱ۳»¿ë (Korean Abstract) |
½º¸¶Æ®ÆùÀÇ ÀåÁ¡Àº ¾ðÁ¦ ¾îµð¼³ª ÀÎÅͳݿ¡ Á¢¼Ó °¡´ÉÇÏ´Ù´Â °Í°ú ´Ã »ç¿ëÀÚ¿Í ÇÔ²² ÀÖ´Ù´Â °ÍÀÌ´Ù. ºñÁî´Ï½ºÀû °üÁ¡¿¡¼ º¸¸é ¾ðÁ¦µç ¼ÒºñÀÚ¿¡°Ô Á¢±ÙÇÒ ¼ö ÀÖ´Â Åë·Î°¡ »ý±ä °ÍÀÌ´Ù. ÇÏÁö¸¸ »ç¿ëÀÚ¿¡°Ô ÇöÀç ÇÏ°í ÀÖ´Â ÀÏÀ» Àá½Ã Áß´ÜÇÏ°í ¿äû¹ÞÀº ÀÏÀ» ÇØ´Þ¶ó°í ¹æÇØÇÒ ¼ö ÀÖ´Â ½ÃÁ¡À» ÄÄÇ»ÅÍ°¡ Á¤È®È÷ ÀÎÁöÇÏ´Â °ÍÀº ¾î·Æ´Ù. ƯÈ÷ ½Å±Ô »ç¿ëÀÚ¿Í °°ÀÌ Ãʱ⠵¥ÀÌÅÍ°¡ ºÎÁ·ÇÑ »óȲ¿¡¼ »ç¿ëÀÚÀÇ Áß´Ü°¡´É¼º (Interruptibility)À» ¿¹ÃøÇÏ´Â °ÍÀº ´õ¿í ¾î·Æ´Ù. º» ¿¬±¸¿¡¼´Â ÀÌ·¯ÇÑ Ãʱ⠵¥ÀÌÅÍ È®º¸°¡ ¾î·Á¿î »óȲ¿¡¼ ±âÁ¸ »ç¿ëÀÚµéÀÇ µ¥ÀÌÅ͸¦ È°¿ëÇÏ¿© ½Å±Ô ½º¸¶Æ®Æù »ç¿ëÀÚÀÇ Áß´Ü°¡´É¼ºÀ» ¿¹ÃøÇÏ´Â ¹æ¹ýÀ» ½ÇÇèÇÏ¿´´Ù. ½Å±Ô »ç¿ëÀÚ¿Í À¯»çÇÑ »ýÈ°ÆÐÅÏÀ» °®´Â ±âÁ¸ »ç¿ëÀÚµéÀ» ã±â À§ÇÏ¿© ½º¸¶Æ®Æù¿¡¼ ÃßÃâ °¡´ÉÇÑ °¢Á¾ ¼¾¼¿Í »ç¿ëÀÚÀÇ Çൿ µ¥ÀÌÅ͸¦ ºÐ¼®ÇÏ°í ºñ±³ÇÏ¿© À¯»ç »ç¿ëÀÚ¸¦ ã´Â ¹æ¹ýÀ» Àû¿ëÇÏ¿´´Ù. ÀÌ ¿¬±¸¸¦ ÅëÇØ ±âÁ¸ »ç¿ëÀÚµéÀÇ µ¥ÀÌÅ͸¦ È°¿ëÇÏ¿© Ãʱ⠵¥ÀÌÅÍ È®º¸°¡ ¾î·Á¿î ½º¸¶Æ®Æù ½Å±Ô »ç¿ëÀÚµéÀÇ Áß´Ü°¡´É¼º ¿¹Ãø Á¤È®µµ¸¦ Çâ»ó½ÃÅ°´Â ¹æ¹ýÀ» Á¦½ÃÇÑ´Ù. |
¿µ¹®³»¿ë (English Abstract) |
One of the advantages of using smart-phones is that it is accessible to the internet wherever the users are and that users carry them most of the time in their daily activities. From a commercial perspective, sellers are able to keep track of users through smart-phones. However, it is not easy to recognize the time when computers ask users to stop what they currently do for a moment, and then ask them to do what they are requested to carry out. In particular, it is challenging to predict interruptibility of users in a circumstance where the amount of dataset is sparse resulting from new users. This research conducted an experiment about a method to predict interruptibility of new smart-phone users especially when it is difficult to acquire the sufficient amount of dataset. It was attempted to analyze various types of sensors and user behaviors extracted from smart-phones to look for the existing users who have a similar activity pattern to the new users. The paper proposes an approach to improve a prediction accuracy of interruptibility of new users whose initial dataset are difficult to obtain by utilizing dataset of the existing users. |
Å°¿öµå(Keyword) |
Áß´Ü°¡´É¼º
»óȲÀÎÁö
½º¸¶Æ®Æù
À¯»çµµ
Interruptibility
Context-Awareness
Smart-phone
Similarity
|
ÆÄÀÏ÷ºÎ |
PDF ´Ù¿î·Îµå
|