Á¤º¸°úÇÐȸ ³í¹®Áö B : ¼ÒÇÁÆ®¿þ¾î ¹× ÀÀ¿ë
Current Result Document : 13 / 13
ÇѱÛÁ¦¸ñ(Korean Title) |
¸ñÀûÁöÇâ ´ëÈ¿¡¼ ÈÀÚ ÀǵµÀÇ Åë°èÀû¿¹Ãø ¸ðµ¨ |
¿µ¹®Á¦¸ñ(English Title) |
A Statistical Prediction Model of Speakers' Intentions in a Goal-Oriented Dialogue |
ÀúÀÚ(Author) |
±èµ¿Çö
±èÇмö
¼Á¤¿¬
Donghyun Kim
Harksoo Kim
Jungyun Seo
|
¿ø¹®¼ö·Ïó(Citation) |
VOL 35 NO. 09 PP. 0554 ~ 0561 (2008. 09) |
Çѱ۳»¿ë (Korean Abstract) |
»ç¿ëÀÚ Àǵµ ¿¹Ãø ±â¼úÀº À½¼ºÀνıâÀÇ Å½»ö °ø°£À» ÁÙÀ̱â À§ÇÑ ÈÄó¸® ¹æ¹ýÀ¸·Î »ç¿ëµÉ ¼ö ÀÖÀ¸¸ç, ½Ã½ºÅÛ Àǵµ ¿¹Ãø ±â¼úÀº À¯¿¬ÇÑ ÀÀ´ä »ý¼ºÀ» À§ÇÑ Àüó¸® ¹æ¹ýÀ¸·Î »ç¿ëµÉ ¼ö ÀÖ´Ù. ÀÌ·¯ÇÑ ½Ç¿ëÀûÀÎ Çʿ伺¿¡ µû¶ó º» ³í¹®¿¡¼´Â ÈÇà°ú °³³ä¿ÀÇ ½ÖÀ¸·Î ÀϹÝÈµÈ ÈÀÚÀÇ Àǵµ¸¦ ¿¹ÃøÇÏ´Â Åë°è ¸ðµ¨À» Á¦¾ÈÇÑ´Ù. ´Ü¼øÇÑ ÈÇà n-±×·¥ Åë°è¸¸À» ÀÌ¿ëÇÑ ±âÁ¸ÀÇ ¸ðµ¨°ú´Â ´Ù¸£°Ô Á¦¾È ¸ðµ¨Àº ÇöÀç ¹ßȱîÁöÀÇ ´ëÈ ÀÌ·ÂÀ» ´Ù¾çÇÑ ¾ð¾î ·¹º§ÀÇ ÀÚÁú ÁýÇÕ(ÈÇà°ú °³³ä¿ ½ÖÀÇ n-±×·¥, ´Ü¼ ´Ü¾î, ¿µ¿ª ÇÁ·¹ÀÓÀÇ »óÅ Á¤º¸)À¸·Î Ç¥ÇöÇÑ´Ù. ±×¸®°í ÃßÃâµÈ ÀÚÁú ÁýÇÕÀ» CRFs(Conditional Random Fields)ÀÇ ÀÔ·ÂÀ¸·Î »ç¿ëÇÏ¿© ´ÙÀ½ ¹ßÈÀÇ Àǵµ¸¦ ¿¹ÃøÇÑ´Ù. ÀÏÁ¤ °ü¸® ¿µ¿ª¿¡¼ ½ÇÇèÀ» ¼öÇàÇÑ °á°ú, Á¦¾È ¸ðµ¨Àº »ç¿ëÀÚÀÇ ÈÇà°ú °³³ä¿ ¿¹Ãø¿¡¼ °¢°¢ 76.25%, 64.21%ÀÇ Á¤È®·üÀ» º¸¿´´Ù. ±×¸®°í ½Ã½ºÅÛÀÇ ÈÇà°ú °³³ä¿ ¿¹Ãø¿¡¼ °¢°¢ 88.11%, 87.19%ÀÇ Á¤È®·üÀ» º¸¿´´Ù. ¶ÇÇÑ ±âÁ¸ ¸ðµ¨°ú ºñ±³ÇÏ¿© 29.32% ³ôÀº Æò±Õ Á¤È®·üÀ» º¸¿´´Ù.
|
¿µ¹®³»¿ë (English Abstract) |
Prediction technique of user's intention can be used as a post-processing method for reducing the search space of an automatic speech recognizer. Prediction technique of system's intention can be used as a pre-processing method for generating a flexible sentence. To satisfy these practical needs, we propose a statistical model to predict speakers' intentions that are generalized into pairs of a speech act and a concept sequence. Contrary to the previous model using simple n-gram statistic of speech acts, the proposed model represents a dialogue history of a current utterance to a feature set with various linguistic levels (i.e. n-grams of speech act and a concept sequence pairs, clue words, and state information of a domain frame). Then, the proposed model predicts the intention of the next utterance by using the feature set as inputs of CRFs (Conditional Random Fields). In the experiment in a schedule management domain, The proposed model showed the precision of 76.25% on prediction of user's speech act and the precision of 64.21% on prediction of user's concept sequence. The proposed model also showed the precision of 88.11% on prediction of system's speech act and the precision of 87.19% on prediction of system's concept sequence. In addition, the proposed model showed 29.32% higher average precision than the previous model.
|
Å°¿öµå(Keyword) |
Àǵµ ¿¹Ãø
ÈÇà ¿¹Ãø
°³³ä¿ ¿¹Ãø
Intention prediction
speech act prediction
concept sequence prediction
|
ÆÄÀÏ÷ºÎ |
PDF ´Ù¿î·Îµå
|