Our system was designed as a therapeutic tool to facilitate the emotional development of children. The proposed system attempts to recognize a child's emotions and estimate their intensity of emotion in a game playing situation. A social agent uses cognitive and non-cognitive factors to estimate the real-time intensity of emotions of the child playing the game. In the system, an autonomus/intelligent agent tries to change the game level according to a child's intensity of emotion for the purpose of encouraging the child to express his/her emotion. This autonomus/intelligent agent uses a robust adaptation model based on the intensity of the child's emotions. Also, an animated pedagogical agent was designed to encourage the child to express emotion using several instructions. The pedagogical agent considers the child's emotions, intensity of emotion, and game states for creating the above instructions. The results showed that the affective gesture recognition model we developed recognized a child's emotion with a considerably high rate of over 82.5%, and the social agent (estimate of intensity of emotion) had a strong relationship with observers' feedback, except in low intensity levels.

View full abstract

This paper deals with the sensibility difference, the characteristics and behavior of drivers who are accustomed to drive on the right-hand side of the road, drive on the left-hand side, and comparison with left-lane drivers. A left-lane driver is defined as a person who is accustomed to drive on the left-hand side of the road. A right-lane driver is defined as a person who is accustomed to drive on the right-hand side of the road. SD (Semantic Differential) method was used for comparing the driving sensibilities of the two types of drivers. GSR (Galvanic Skin Response) was measured to obtain changes in the physiological signal in terms of objective evaluation while driving on simulated driving. Driving mental workload was also evaluated based on MNASA-TLX (Modified NASA-TLX) as a subjective evaluation. Results indicate that the sensibility and characteristics of the right-lane drivers were shown different in comparison with those of the left-lane drivers when driving on the opposite lane due to their habit and traits. The result of SD method is that right-lane drivers presented a feeling of incompatibility by the road convention, left-lane drivers, however, presented a feeling from driving simulator as the first factor. In the result of mental workload, the right-lane drivers were in higher stress condition than left-lane driver due to driving on opposite lane that they were not used to. Consequently, driving habit has influence on driver's sensibility and mental workload.

View full abstract

In general, the methods for measuring human emotion are divided in two, one is the psychological way based on user's subjective evaluation and the other is the physiological way based on physiological signals. However, these methods have some limitations. Therefore, in this study we suggested the methods for measuring user's emotion in the natural and accessible environment for the design field. In order to observe user's emotional changes while they interact with a product, we have extracted some emotional words and representative emotions, and made a set of subjective evaluation scales. With these scales, emotion logging program “Video TAME”was developed as an effective measurement tool for complementing the current psychological methods. In the Testing Module of VideoTAME, subjects evaluate their emotional changes through playing and watching the video clips of their performing tasks in the experiment room. In the Analyzing Module, the researchers replay the results created by subjects during the experiment and analyze the results using Microsoft Excel. It is hoped that this study will be of great help for designers for effective measurement of user's emotions naturally expressed while using a product.

View full abstract

Townscape colours have been a main factor in urban development. For townscape colours, keeping colour harmony within the environment is a common goal. In addition to the colour harmony, expressing characteristics and impressions of the town in townscape colours are other meaningful goals. This paper describes IroKage, a colour planning support system for improving townscapes. The system offers some colour combination proposals based on three elements: colour harmony, impressions of the townscape, and cost for the change of colours. First, two evaluation models to quantify colour harmony and impression of the townscape are developed from the approach of colour harmony principles and Kansei engineering, respectively. Next, the system is constructed using an evolutionary algorithm and the two evaluation models. After the construction, performance tests are conducted. The results show that the system has achieved sufficient ability to propose appropriate colour combinations with minimum colour changes.

View full abstract

KANSEI can be defined as the human ability to translate stimulus into impression. When the KANSEI is defined as ability, KANSEI ability is varied with individuals. Actually, some subjects may reproduce to translate stimulus into perceptions, but another subjects may not; while some subjects may translate some perceptions into various impressions, but another subjects may not. As a result, KANSEI data includes individual differences. If the variety of the individual differences on these translations can be measured as KANSEI ability, it can be helpful for analyzing the structural model of KANSEI evaluation. In this paper, it is supposed that a hierarchy structural model, which includes the perceptual stage, image stage and evaluation stage. Indices to measure translation abilities from stimulus into perceptual stage and from the perceptual stage into the image stage are proposed using the Semantic Differential (SD) method. KANSEI Quality of hubcap designs is included as a case study, to which the proposed indices are applied.

View full abstract

The visual field test can be used to examine visual field impairment or diagnose the progression of ophthalmologic diseases such as glaucoma. Visual field impairment is related to optic nerve impairment. The doctor's views of optic nerve impairments are based on optic discs and retinal nerve fiber layers. In this research, we developed glaucoma diagnosis software, applied a semantic differential method, and calculated the glaucoma probability using multivariate analysis. The software is capable of receiving perimetric Humphrey perimeter data. Visual field tests were performed on 7 eyes of 7 patients with glaucoma, and we calculated their glaucoma probability. The fundus photography was performed in 6 eyes of 6 healthy persons. We used the software to measure the number of pixels in an image, and it also calculated their cup/disc ratios, the size and area of the disc and cup, and so on. We surveyed the correlation between glaucoma probability and the Collaborative Initial Glaucoma Treatment Study (CIGTS). The correlation coefficient was 0.881, which shows significant correlation between these two factors. The vertical cup/disc ratio was 0.79±0.04, and the horizontal cup/disc ratio 0.75±0.03. The disc area was 28031±4946, and that of the cup was 16075±2383: the average ±standard deviation. We established a questionnaire dialog using the semantic differential method and adopted scrollbars to the questionnaire dialog to reflect the doctor's delicate meanings.

View full abstract

This paper discusses non-linearity in KANSEI evaluation data and attempts to better understand non-linear relationships between design attributes. Conventionally, the basis of KANSEI analysis has been multivariate analyses and linear modeling. These methods have proven unreliable and problematic. Non-linearity may be typically represented by the interaction between design elements, which often is represented as multicollinearity. To better understand such interactions, a method is proposed that expresses the effect of combining multiple design elements, and uses 'f-then' rules in conjunction with decision trees. The method is also described for constructing these decision trees by using genetic algorithms (GA). Our concern is to consider the adequateness of the proposed method for analyses of non-linear KANSEI data. Two actual examples are shown in order to investigate this problem. An example of KANSEI experiment on interior automobile space is used to examine divided partitions as tree representations. Another illustration of canned coffee design shows a comparison between the results of our method and quantification type I. We can conclude from these examples such that the proposed method indicates in more detailed rules and a deeper understanding of information extracted from KANSEI evaluation data.

View full abstract