µ¥ÀÌÅͺ£À̽º ¿¬±¸È¸Áö(SIGDB)
Current Result Document : 6 / 6
ÇѱÛÁ¦¸ñ(Korean Title) |
Ŭ·¯½ºÅÍ ±â¹Ý Å°¿öµå ¿¬°ü¸ÁÀÇ ÀÚµ¿ ±¸Ãà |
¿µ¹®Á¦¸ñ(English Title) |
Automatic construction of cluster-based keyword association network |
ÀúÀÚ(Author) |
À¯Çѹ¬
±èÇÑÁØ
Han-Mook Yoo
Han-joon Kim
|
¿ø¹®¼ö·Ïó(Citation) |
VOL 33 NO. 01 PP. 0015 ~ 0025 (2017. 04) |
Çѱ۳»¿ë (Korean Abstract) |
º» ³í¹®Àº ±âÁ¸ÀÇ TextRank ¾Ë°í¸®Áò¿¡ »óÈ£Á¤º¸·® ôµµ¸¦ °áÇÕÇÏ¿© ±ºÁý ±â¹Ý¿¡¼ Å°¿öµå ÃßÃâÇÏ´Â ClusterTextRank ±â¹ý°ú ÃßÃâµÈ Å°¿öµå¸¦ ÃÖ¼Ò½ÅÀåÆ®¸®¸¦ ÀÌ¿ëÇÑ ¿¬°ü¸Á ±¸Ãà ±â¹ýÀ» Á¦¾ÈÇÑ´Ù. Á¦¾È ±â¹ýÀº k-means ±ºÁýÈ ¾Ë°í¸®ÁòÀ» ÀÌ¿ëÇÏ¿© ¹®¼µéÀ» ¿©·¯ ±ºÁýÀ¸·Î ³ª´©°í, °¢ ±ºÁý¿¡ Æ÷ÇÔµÈ ´Ü¾îµéÀ» ÃÖ¼Ò½ÅÀåÆ®¸® ±×·¡ÇÁ·Î Ç¥ÇöÇÑ ÈÄ ÀÌ¿¡ ±Ù°ÅÇÑ ±ºÁý Á¤º¸·®À» °í·ÁÇÏ¿© Å°¿öµå¸¦ ÃßÃâÇÑ´Ù. ±× ´ÙÀ½ ÃßÃâµÈ Å°¿öµåµé°£¿¡ À¯»çµµ¸¦ °è»êÇÑ ÈÄ ÃÖ¼Ò½ÅÀåÆ®¸® ±×·¡ÇÁ·Î Ç¥ÇöÇÏ°í, À̸¦ Å°¿öµå ¿¬°ü¸ÁÀ¸·Î È°¿ëÇÑ´Ù. Á¦¾È ±â¹ýÀÇ ¼º´ÉÀ» Æò°¡Çϱâ À§ÇØ ¿©Çà °ü·Ã ºí·Î±× µ¥ÀÌÅ͸¦ ÀÌ¿ëÇÏ¿´À¸¸ç, Á¦¾È ±â¹ýÀÌ ±âÁ¸ TextRank ¾Ë°í¸®Áòº¸´Ù Å°¿öµå ÃßÃâÀÇ Á¤È®µµ°¡ ¾à 12% °¡·® °³¼±µÊÀ» º¸ÀδÙ.
|
¿µ¹®³»¿ë (English Abstract) |
In this paper, we propose a novel way of producing keyword networks, named ClusterTextRank, which extracts significant key words from a set of clusters with mutual information metric, and constructs an association network from their minimal spanning tree. The proposed method decomposes documents into multiple clusters through the k-means clustering, and expresses the words within each cluster as a minimum spanning tree graph. The significant key words are determined by evaluating their mutual information within clusters. Then, the method calculates the similarity among the extracted key words, and the results are represented as a minimum spanning tree, called a keyword association network. To evaluate the performance of the proposed method, we used travel-related blog data and showed that the proposed method outperforms the existing TextRank algorithm by about 12% in terms of accuracy.
|
Å°¿öµå(Keyword) |
»óÈ£Á¤º¸·®
ÃÖ¼Ò½ÅÀåÆ®¸®
Ŭ·¯½ºÅ͸µ
Å°¿öµåÃßÃâ
ÅؽºÆ®¸¶ÀÌ´×
Mutual Information
Minimal Spanning Tree
Clustering
Keyword extraction
Text mining
|
ÆÄÀÏ÷ºÎ |
PDF ´Ù¿î·Îµå
|