Á¤º¸Ã³¸®ÇÐȸ ³í¹®Áö ¼ÒÇÁÆ®¿þ¾î ¹× µ¥ÀÌÅÍ °øÇÐ
Current Result Document :
ÇѱÛÁ¦¸ñ(Korean Title) |
±¹¹Îû¿ø ÁÖÁ¦ ºÐ¼® ¹× µö·¯´× ±â¹Ý ´äº¯ °¡´É û¿ø ¿¹Ãø |
¿µ¹®Á¦¸ñ(English Title) |
Topic Analysis of the National Petition Site and Prediction of Answerable Petitions Based on Deep Learning |
ÀúÀÚ(Author) |
¼ÛÁØÈ£
±è¼º¼ö
Àü¹®¼®
Jun Ho Song
Sung-Soo Kim
Moon-Seog Jun
¿ìÀ±Èñ
±èÇöÈñ
Woo Yun Hui
Hyon Hee Kim
|
¿ø¹®¼ö·Ïó(Citation) |
VOL 09 NO. 02 PP. 0045 ~ 0052 (2020. 02) |
Çѱ۳»¿ë (Korean Abstract) |
û¿Í´ë ±¹¹Î û¿ø »çÀÌÆ®°¡ °³¼³µÈ ÀÌ·¡·Î ¸¹Àº °ü½ÉÀ» ¹Þ°í ÀÖ´Ù. º» ³í¹®¿¡¼´Â ±¹¹Î û¿øÀÇ ÁÖÁ¦¸¦ ºÐ¼®ÇÏ°í µö·¯´×À» È°¿ëÇÏ¿© ´äº¯ °¡´ÉÇÑ Ã»¿øÀ» ¿¹ÃøÇÏ´Â ¸ðµ¨À» Á¦¾ÈÇÏ¿´´Ù. ¸ÕÀú, Ãßõ¼øÀ¸·Î 1,500°³ÀÇ Ã»¿ø±ÛÀ» ¼öÁýÇÏ¿´°í, K-means Ŭ·¯½ºÅ͸µÀ» Àû¿ëÇÏ¿© û¿ø±ÛÀ» ±ºÁýÇÏ¿© ´ëÁÖÁ¦¸¦ Á¤ÀÇÇÏ°í, º¸´Ù ±¸Ã¼ÀûÀÎ ¼¼ºÎ ÁÖÁ¦¸¦ Á¤ÀÇÇϱâ À§È÷¿© ÅäÇÈ ¸ðµ¨¸µÀ» ½Ç½ÃÇÏ¿´´Ù. ´ÙÀ½À¸·Î´Â LSTMÀ» È°¿ëÇÑ ´äº¯ °¡´ÉÇÑ Ã»¿ø ¿¹Ãø ¸ðµ¨À» »ý¼ºÇÏ¿©, 20¸¸ÀÇ Ã»¿øµ¿ÀǸ¦ ¾ò´Â û¿øÀ» ¿¹ÃøÇϱâ À§ÇÑ ¸ðµ¨À» °³¹ßÇÏ¿´´Ù. À̸¦ À§ÇØ ±ÛÀÇ ÁÖÁ¦¿Í º»¹®»Ó¸¸ ¾Æ´Ï¶ó ±ÛÀÇ ±æÀÌ, Ä«Å×°í¸®, ƯÁ¤ Ç°»çÀÇ ºñÀ²ÀÌ ¿µÇâÀ» ¹ÌÄ¥ ¼ö ÀÖ´ÂÁö¸¦ »ìÆ캸¾Ò´Ù. ±× °á°ú, º»¹®°ú ÇÔ²² ±ÛÀÇ ±æÀÌ, Ä«Å×°í¸®, ü¾ð, ¿ë¾ð, µ¶¸³¾ð, ¼ö½Ä¾ðÀÇ Ç°»çÀÇ ºñÀ²À» º¯¼ö·Î Ãß°¡ÇÑ ¸ðµ¨ÀÇ f1-score°¡ 0.9 ÀÌ»óÀ¸·Î ±ÛÀÇ Á¦¸ñ°ú º»¹®À» º¯¼ö·Î ÇÏ´Â ¸ðµ¨º¸´Ù ¿¹Ãø·ÂÀÌ ³ôÀ½À» ¾Ë ¼ö ÀÖ¾ú´Ù.
|
¿µ¹®³»¿ë (English Abstract) |
Since the opening of the national petition site, it has attracted much attention. In this paper, we perform topic analysis of the national petition site and propose a prediction model for answerable petitions based on deep learning. First, 1,500 petitions are collected, topics are extracted based on the petitions¡¯ contents. Main subjects are defined using K-means clustering algorithm, and detailed subjects are defined using topic modeling of petitions belonging to the main subjects. Also, long short-term memory (LSTM) is used for prediction of answerable petitions. Not only title and contents but also categories, length of text, and ratio of part of speech such as noun, adjective, adverb, verb are also used for the proposed model. Our experimental results show that the type 2 model using other features such as ratio of part of speech, length of text, and categories outperforms the type 1 model without other features.
|
Å°¿öµå(Keyword) |
Diffie-Hellman
Asymmetric Key Cryptography
Key Exchange
µðÇÇ-Ç︸
ºñ´ëĪ Å° ¾ÏÈ£È
Å° ±³È¯ ¾Ë°í¸®Áò
National Petition
Topic Analysis
Topic Modeling
K-means Clustering
LSTM
Deep Learning
±¹¹Îû¿ø
ÁÖÁ¦ ºÐ¼®
ÅäÇÈ ¸ðµ¨¸µ
K-means Ŭ·¯½ºÅ͸µ
LSTM
µö·¯´×
|
ÆÄÀÏ÷ºÎ |
PDF ´Ù¿î·Îµå
|