Seite - 153 - in Intelligent Environments 2019 - Workshop Proceedings of the 15th International Conference on Intelligent Environments

In recent years, there has been a growing demand for awareness of depression among emotions and depression is considered to affect physical bodies [3,4]. Therefore, we need a system to recognize the depression. Although there are other techniques such as image processing, we consider utilization of electroencephalographs (EEGs) due to its advantage of low cost. As a advantage of emotion recognition by EEG, it can be used even if it is not in front of the device with a camera such as a personal computer, and not only emotion recognition but also other recognition such as concentration degree and stress can be done, and the goodness of the affinity of the headset and the electroencephalogram, It is possible to use the call center where the change of feelings is considered a problem, and to study. Girardi showed that emotional recognition in EEG is effective in various biological signals [5]. It is also an effective tool for detecting dementia, alcoholism, autism spectrum, etc. [6] [7] [8]. There are various kinds of emotion recognition methods such as image recognition of face [9] [10], speech [11], heart rate [12], EEGs, but in this study we decided to treat emotion recognition by EEG. EEG is more expensive than the number of electrodes, and there are problems such as restricting the motion. Moreover, there is a problem that the more electrodes take longer to install. There are various types of EEG, such as expensive ones with a large number of electrodes, inexpensive ones with low number of electrodes, and fewer electrodes, but the arrangement of electrodes can be changed. Individual differences may appear in EEG due to the shape of the skull, sebum on the scalp, and the presence of hair.[13] The characteristic of EEG is that the information that can be obtained by the electrode site of the electroencephalogram is different. Therefore, the purpose of this study was to investigate the combination of the drop and the high contribution electrodes in the emotion recognition in EEG. What type of electroencephalogram should be used to investigate the depression using a small number of electrodes, if the electroencephalogram that can change the electrode arrangement, we thought that it might be an index that it was easy to recognize the drop by the arrangement of the electrode. In this paper, we explain a method to reduce electrodes in emotional recognition of depression by using an EEG from the EEG data of DEAP. As a result, it was confirmed that the multiple correlation coefficient of EEG data and depression increased as the number of electrodes increased. The relationship between EEG data acquired from the electrode site selected by the method in this research and the feeling of depression was clarified. The rest of the paper is organized as follows. Section 2 includes a brief description of the MWM system architecture and how to make feature quantities and investigation by LARS. Section 3 describes the evaluation result and research. Section 4 presents the most important conclusions and provides directions for additional research. 2. Implementation of MWM system In this chapter, we will describe how to proceed with the analysis in this research. Figure 1 shows an overview of how to proceed with analysis. Y.Tsurugasaki etal. / IdentificationofEffectiveEEGElectrodes forDepressionSensing 153