Á¤º¸Ã³¸®ÇÐȸ ³í¹®Áö ¼ÒÇÁÆ®¿þ¾î ¹× µ¥ÀÌÅÍ °øÇÐ
Current Result Document : 268 / 269
ÇѱÛÁ¦¸ñ(Korean Title) |
¾ÆÆÄÆ® ÇÏÀÚ º¸¼ö ½Ã¼³°ø»ç ¼¼ºÎ°øÁ¾ ¸Ó½Å·¯´× ºÐ·ù ½Ã½ºÅÛ¿¡ °üÇÑ ¿¬±¸ |
¿µ¹®Á¦¸ñ(English Title) |
Classifying Sub-Categories of Apartment Defect Repair Tasks: A Machine Learning Approach |
ÀúÀÚ(Author) |
±èÀºÇý
ÁöÈ«±Ù
±èÁö³ª
¹ÚÀºÀÏ
¾öÀç¿ë
Eunhye Kim
HongGeun Ji
Jina Kim
Eunil Park
Jay Y. Ohm
|
¿ø¹®¼ö·Ïó(Citation) |
VOL 10 NO. 09 PP. 0359 ~ 0366 (2021. 09) |
Çѱ۳»¿ë (Korean Abstract) |
´ëÇѹα¹ °Ç¼³»çµéÀº ¾ÆÆÄÆ® ÇÏÀÚ Á¤º¸¸¦ ÃàÀûÇÏ°í º¸¼öÀÛ¾÷À» °ü¸®Çϱâ À§ÇÑ ½Ã½ºÅÛÀ» ¿î¿µÇϴµ¥ »ó´çÇÑ Àη°ú ºñ¿ëÀ» ÅõÀÚÇÏ°í ÀÖ´Ù. º» ¿¬±¸¿¡¼´Â ÇÏÀÚ Á¢¼ö »ó¼¼³»¿ë ÅؽºÆ® µ¥ÀÌÅ͸¦ ÀÌ¿ëÇÏ¿© ÇÏÀÚ º¸¼ö ½Ã¼³°ø»ç¿¡ µû¸¥ ¼¼ºÎ°øÁ¾À» ºÐ·ùÇÏ´Â ¸Ó½Å·¯´× ¸ðµ¨À» Á¦¾ÈÇÑ´Ù. µÎ °¡Áö ´Ü¾î ÀÓº£µù(Bag-of-words, Term Frequency-Inverse Document Frequency (TF-IDF))°ú µÎ °¡Áö ºÐ·ù±â(Support Vector Machine, Random Forest)¸¦ ÅëÇØ Çѱ¹¾î·Î ÀÛ¼ºµÈ 65¸¸°Ç ÀÌ»óÀÇ ÇÏÀÚ Á¢¼öµ¥ÀÌÅͷκÎÅÍ ÇÏÀÚº¸¼ö ½Ã¼³°ø»ç ¼¼ºÎ°øÁ¾À» ºÐ·ùÇß´Ù. ƯÈ÷, À̹ø ¿¬±¸¿¡¼´Â ƯÁ¤ ½Ã¼³°ø»ç(¸¶°¨°ø»ç)ÀÇ 9°³ ¼¼ºÎ°øÁ¾(°¡ÀüÁ¦Ç°, µµ¹è°ø»ç, µµÀå°ø»ç, ¹ÌÀå°ø»ç, ¼®°ø»ç, ¼öÀå°ø»ç, ¿Á³»°¡±¸°ø»ç, ÁÖ¹æ±â±¸°ø»ç, ŸÀÏ°ø»ç)À» ºÐ·ùÇÏ´Â ÀÌÁø ºÐ·ù ¸ðµ¨°ú ´ÙÁß ºÐ·ù ¸ðµ¨À» ¿¬±¸Çß´Ù. ±× °á°ú, TF-IDF¿Í Random Forest¸¦ »ç¿ëÇÑ µÎ°¡Áö ºÐ·ù ¸ðµ¨¿¡¼ 90%ÀÌ»óÀÇ Á¤È®µµ, Á¤¹Ðµµ, ÀçÇöÀ² ¹× F1Á¡¼ö¸¦ È®ÀÎÇß´Ù. |
¿µ¹®³»¿ë (English Abstract) |
A number of construction companies in Korea invest considerable human and financial resources to construct a system for managing apartment defect data and for categorizing repair tasks. Thus, this study proposes machine learning models to automatically classify defect complaint text-data into one of the sub categories of ¡®finishing work¡¯ (i.e., one of the defect repair tasks). In the proposed models, we employed two word representation methods (Bag-of-words, Term Frequency-Inverse Document Frequency (TF-IDF)) and two machine learning classifiers (Support Vector Machine, Random Forest). In particular, we conducted both binary- and multi- classification tasks to classify 9 sub categories of finishing work: home appliance installation work, paperwork, painting work, plastering work, interior masonry work, plaster finishing work, indoor furniture installation work, kitchen facility installation work, and tiling work. The machine learning classifiers using the TF-IDF representation method and Random Forest classification achieved more than 90% accuracy, precision, recall, and F1 score. We shed light on the possibility of constructing automated defect classification systems based on the proposed machine learning models. |
Å°¿öµå(Keyword) |
¾ÆÆÄÆ®
ÇÏÀÚ
½Ã¼³°ø»ç
¼¼ºÎ°øÁ¾
¸¶°¨°ø»ç
¸Ó½Å·¯´×
Apartment
Defect
Repair Tasks
Sub Category
Finishing Works
Machine Learning
|
ÆÄÀÏ÷ºÎ |
PDF ´Ù¿î·Îµå
|