• Àüü
  • ÀüÀÚ/Àü±â
  • Åë½Å
  • ÄÄÇ»ÅÍ
´Ý±â

»çÀÌÆ®¸Ê

Loading..

Please wait....

±¹³» ³í¹®Áö

Ȩ Ȩ > ¿¬±¸¹®Çå > ±¹³» ³í¹®Áö > Çѱ¹ÀÎÅͳÝÁ¤º¸ÇÐȸ ³í¹®Áö

Çѱ¹ÀÎÅͳÝÁ¤º¸ÇÐȸ ³í¹®Áö

Current Result Document :

ÇѱÛÁ¦¸ñ(Korean Title) ±â°èÇнÀ ¾Ë°í¸®Áò¿¡ ±â¹ÝÇÑ ³úÆÄ µ¥ÀÌÅÍÀÇ °¨Á¤ºÐ·ù ¹× Á¤È®µµ Çâ»ó¿¡ °üÇÑ ¿¬±¸
¿µ¹®Á¦¸ñ(English Title) A research on the emotion classification and precision improvement of EEG(Electroencephalogram) data using machine learning algorithm
ÀúÀÚ(Author) ÀÌÇöÁÖ   ½Åµ¿ÀÏ   ½Åµ¿±Ô   Hyunju Lee   Dongil Shin   Dongkyoo Shin  
¿ø¹®¼ö·Ïó(Citation) VOL 20 NO. 05 PP. 0027 ~ 0036 (2019. 10)
Çѱ۳»¿ë
(Korean Abstract)
º» ¿¬±¸¿¡¼­´Â °ø°³µÈ ³úÆÄ µ¥ÀÌÅÍÀÎ DEAP(A Database for Emotion Analysis using Physiological Signals) µ¥ÀÌÅÍ ¼¼Æ®¸¦ È°¿ëÇÑ °¨Á¤ºÐ·ù¡¤ºÐ¼® ¹× Á¤È®µµ Çâ»ó¿¡ ´ëÇÑ ½ÇÇèÀ» ÁøÇàÇÏ¿´´Ù. ½ÇÇè¿¡´Â 32¸í¿¡ ´ëÇÑ 32°³ÀÇ ³úÆÄÃøÁ¤ ä³Î µ¥ÀÌÅÍ°¡ ¸ðµÎ »ç¿ëµÇ¾ú´Ù. Àü󸮰úÁ¤¿¡¼­´Â ³úÆÄ µ¥ÀÌÅÍ¿¡ ´ëÇÑ 256Hz »ùÇøµÀÛ¾÷À» ÁøÇàÇÏ¿´°í, À¯ÇÑ ÀÓÆÞ½º ÀÀ´ä ÇÊÅ͸¦ »ç¿ëÇÏ¿© ÁÖÆļö ´ë¿ªº°·Î ½êŸ(4-8Hz), ½½·Î ¾ËÆÄ(8-10Hz), ¾ËÆÄ (8-12Hz), º£Å¸(12-30Hz), °¨¸¶(31-45Hz) ÆÄÇü¿¡ ´ëÇÑ µ¥ÀÌÅ͸¦ ÃßÃâÇÏ¿´´Ù. ÃßÃâÇÑ µ¥ÀÌÅÍ´Â ½Ã°£-ÁÖÆļö º¯ÇüÀ» ÅëÇÏ¿© µ¥ÀÌÅÍÀÇ »óŸ¦ ±¸ºÐÇÑ ÈÄ¿¡, µ¶¸³¼ººÐºÐ¼®¹æ¹ýÀ» ÅëÇØ ÀâÀ½(Artifact)À» Á¦°ÅÇÏ¿© µ¥ÀÌÅ͸¦ Á¤Á¦Çß´Ù. µµÃâµÈ µ¥ÀÌÅÍ´Â ºÐ·ù±â ±â°èÇнÀ ¾Ë°í¸®Áò ½ÇÇèÀ» ½ÃÇàÇÒ ¼ö ÀÖµµ·Ï CSV ÆÄÀÏ·Î º¯Çü ÇÏ¿´À¸¸ç, °¨Á¤ºÐ·ù¿¡´Â Arousal-Valence Æò¸éÀ» »ç¿ëÇÏ¿´´Ù. °¨Á¤Àº ¡°±àÁ¤Àû(Positive)¡±, ¡°ºÎÁ¤Àû(Negative)¡± ÀÌ¿Ü¿¡ Æò¿ÂÇÑ »óÅ·ΠÁ¸ÀçÇÏ´Â ¡°Á߸³Àû(Neutral)¡±ÀÇ 3°¡Áö »óÅ·ΠºÐ·ùÇÏ¿´´Ù. Á¤È®µµ¸¦ °³¼±Çϱâ À§Çؼ­ ·£´ý Æ÷·¹½ºÆ®(Random Forest) ¾Ë°í¸®Áò¿¡ ¼Ó¼º ¼±ÅÃÀû ºÐ·ù±â(Attribute Selected Classifier: ASC) ¹æ½Ä¿¡ ÀÇÇØ ¼±ÅÃµÈ ¼Ó¼ºÀ» Àû¿ëÇÏ¿© ½ÇÇèÇÏ¿´´Ù. Á¤È®µµ´Â ¡°°¢¼º(Arousal)¡± ºÎºÐ¿¡¼­ KoelstraÀÇ °á°úº¸´Ù ¡°32.48%¡± ³ôÀº °á°ú°¡ µµÃâµÇ¾ú°í, LiuÀÇ ½ÇÇèÀÇ ¡°Á¤¼­°¡(Valence)¡±¿Í ºñ±³Çغ¸¸é ASC(Random Forest) °á°ú°¡ ¡°8.13%¡± ´õ ³ôÀº °á°ú¸¦ µµÃâÇÏ¿´´Ù. Á¤È®µµ¸¦ °³¼±Çϱâ À§ÇØ ASC ¹æ½ÄÀ» Àû¿ëÇÑ ·£´ý Æ÷·¹½ºÆ® ºÐ·ù±â ½ÇÇè°á°ú¿¡¼­´Â ÀüüÆò±ÕÀ» ±âÁØÀ¸·Î ±âÁ¸ ¿¬±¸ °á°ú¿Í ´ëºñÇÏ¿© ¡°2.68%¡± ³ôÀº Á¤È®µµ°¡ µµÃâµÇ¾ú´Ù.
¿µ¹®³»¿ë
(English Abstract)
In this study, experiments on the improvement of the emotion classification, analysis and accuracy of EEG data were proceeded, which applied DEAP (a Database for Emotion Analysis using Physiological signals) dataset. In the experiment, total 32 of EEG channel data measured from 32 of subjects were applied. In pre-processing step, 256Hz sampling tasks of the EEG data were conducted, each wave range of the frequency (Hz); Theta, Slow-alpha, Alpha, Beta and Gamma were then extracted by using Finite Impulse Response Filter. After the extracted data were classified through Time-frequency transform, the data were purified through Independent Component Analysis to delete artifacts. The purified data were converted into CSV file format in order to conduct experiments of Machine learning algorithm and Arousal-Valence plane was used in the criteria of the emotion classification. The emotions were categorized into three-sections; ¡®Positive¡¯, ¡®Negative¡¯ and ¡®Neutral¡¯ meaning the tranquil (neutral) emotional condition. Data of ¡®Neutral¡¯ condition were classified by using Cz(Central zero) channel configured as Reference channel. To enhance the accuracy ratio, the experiment was performed by applying the attributes selected by ASC(Attribute Selected Classifier). In ¡°Arousal¡± sector, the accuracy of this study¡¯s experiments was higher at ¡°32.48%¡± than Koelstra¡¯s results. And the result of ASC showed higher accuracy at ¡°8.13%¡± compare to the Liu¡¯s results in ¡°Valence¡±. In the experiment of Random Forest Classifier adapting ASC to improve accuracy, the higher accuracy rate at ¡°2.68%¡± was confirmed than Total mean as the criterion compare to the existing researches.
Å°¿öµå(Keyword) DEAP µ¥ÀÌÅÍ ¼¼Æ®   µ¶¸³¼ººÐºÐ¼®   Arousal-Valence Æò¸é   ·£´ý Æ÷·¹½ºÆ®   ¼Ó¼º ¼±ÅÃÀû ºÐ·ù±â   DEAP dataset   ICA   Arousal-Valence plane   Random Forest   Attribute Selected Classifier  
ÆÄÀÏ÷ºÎ PDF ´Ù¿î·Îµå