日本感性工学会論文誌
Online ISSN : 1884-5258
Print ISSN : 1884-0833
最新号
選択された号の論文の9件中1~9を表示しています
原著論文
  • 加藤 貴司, 竹内 志津江, 橋本 稔
    2018 年 17 巻 5 号 p. 515-521
    発行日: 2018年
    公開日: 2018/12/26
    [早期公開] 公開日: 2018/11/20
    ジャーナル フリー
    In this research, we developed a flexible and easy to wear pants type robotic wear as a new version of non-exoskeleton type robotic wear curara®. Wearable robots are being developed to support independence of care receivers. One of the major problems of existing wearable robots is that it takes long time to wear. Therefore, we aimed to develop a robot wear that can be worn by a user in a short time with no assistance. We achieved this target by combining fiber material with motor power section. Performances of both the developed robotic wear and an existing robotic wear were compared. It was found that the developed robotic wear was able to shorten the wearing time by about 44% while satisfying power assistance and other requirements.
  • 中村 充志, 瀧澤 生, 星 泰成, 綱島 秀樹, 陳 キュウ
    2018 年 17 巻 5 号 p. 523-529
    発行日: 2018年
    公開日: 2018/12/26
    [早期公開] 公開日: 2018/10/23
    ジャーナル フリー
    In this paper, we propose an automatic font generation algorithm which can accurately represent Kansei of the input image. Provided an arbitrary image, a corresponding font is generated according to the emotion information included in the image. The proposed method consists of two stages: learning stage and generation stage. In the first stage, the learning unit learns multiple fonts by using the image generation model “zi2zi” (meaning characters to characters), which is based on the GAN principle. In the generation stage, the score of the emotion information is obtained using “Emotion API” for the analysis of the image, then an inference is made by using the weights of the Generator model obtained by the learning, and the font is generated finally. Evaluation experiments via web questionnaires show that the fonts generated by the proposed algorithm correctly reflect emotions, demonstrating the effectiveness of our algorithm.
  • 小暮 枝里子, 齊藤 史哲, 石津 昌平
    2018 年 17 巻 5 号 p. 537-545
    発行日: 2018年
    公開日: 2018/12/26
    [早期公開] 公開日: 2018/11/20
    ジャーナル フリー
    Analysis technology of document data on customer review has been received broad attention because it can be reflected in improving products and services. It is difficult to read all of the text data, and classification by learning model is effective for acquiring knowledge from the voices of customers. In constructing a learning model for large-scale data, it is important to make effective use of data without labels. In this situation semi-supervised learning has gained a lot of success. The purposes of this research are to reduce the cost of labeling document data by human and to improve prediction accuracy of learning models. In this research, we focused on classification task of customer review data with limited labeling. Through extending the random forest to semi supervised learning, we achieved improved classification accuracy of customer review data.
  • 井上 征矢
    2018 年 17 巻 5 号 p. 547-556
    発行日: 2018年
    公開日: 2018/12/26
    [早期公開] 公開日: 2018/11/20
    ジャーナル フリー
    Since the meanings of Japanese words and sign language words do not necessarily correspond to each other, sometimes a single sign language is used to express distinct concepts in the Japanese language. This study focused on the concepts used in the color education that are difficult to distinguish in sign language and discussed a method to express them distinctively. To this end, we focused on terms that describe color impressions; then,we conducted a survey to understand how hearing-impaired students actually express those terms. The results showed various tendencies including the following: 1) several terms were expressed differently depending on individual students, 2) the same sign has been used for different meanings depending on the student, 3) many students used the same sign to express two different concepts that can be distinguished in the Japanese language, such as “koi” (deep) and “hakkiri” (distinctly), “azayaka” (vivid) or “sunda” (clear) and “kirei” (beautiful), and “nigotta” (muddy) and “kitanai” (dirty).
  • 宮下 達哉, 木村 敦, 岡 隆
    2018 年 17 巻 5 号 p. 561-566
    発行日: 2018年
    公開日: 2018/12/26
    [早期公開] 公開日: 2018/12/14
    ジャーナル フリー
    Several studies have reported the factors which determine individual differences in aesthetic evaluations of visual arts. Our previous studies suggested that the aesthetic dimension of value (ADV) was a crucial factor relating to the individual differences in aesthetic evaluations. However, the paintings used in our previous studies were selected from the masterpieces painted by famous artists. The present study explored whether the ADV would relate to aesthetic evaluations for both good and bad arts. Undergraduates (N = 166) were asked to rate 14 paintings (including seven good arts and seven bad arts selected from the Museum of Bad Art) on four scales of aesthetic evaluation and to complete a questionnaire assessing their degree of the ADV. The results demonstrated that the ADV related to aesthetic evaluations for both good and bad arts. These results suggest that the ADV related to the subjective value of the paintings regardless of their reputation.
  • 山田 篤拓, 橋本 翔, 長田 典子
    2018 年 17 巻 5 号 p. 567-576
    発行日: 2018年
    公開日: 2018/12/26
    [早期公開] 公開日: 2018/12/14
    ジャーナル フリー
    To meet user's affective needs, a number of efforts have been addressed to index kansei. However, the conventional method based on subjective evaluation experience take much time and effort to experiment and analyze. To solve this problem, we propose a method that automatically indexes impressions which is the first phase of attempting to index kansei based on evaluative expression dictionaries from the review data. This method consists of three steps. First, we collect evaluation words and categorize them as impression or emotion using word classes and the evaluative expression dictionary. Second, we extract the impression topics using the topic model, which uses only the evaluation words of impressions. Finally, we score each product for each impression topic, using the frequencies of evaluation words and term-scores. The results of an application of the method to the review texts of wristwatches and the subjective evaluation experiment show the validity of our method.
  • 高橋 正人, 大谷 毅
    2018 年 17 巻 5 号 p. 577-585
    発行日: 2018年
    公開日: 2018/12/26
    [早期公開] 公開日: 2018/12/21
    ジャーナル フリー
    In our previous paper, we proposed diffusion equation model in behavioral space in order to explain the relationship between power of items and selling power. Behavioral space is the set composed of personal figures in various space of human behavior. Responsible designer of fashion apparels designs items of his brand by extracting subset composed of personal figures of clients of his brand. This study aims to elucidate the dimension of image valuables set up by responsible designer in behavioral space based on the diffusion model described in our previous paper. If the dimension is 1-dimension, it is shown that the total sales numbers of apparel change proportional to the root of time t. On the other hand, it is also shown that the total sales numbers of apparel change proportional to t and t1.5 in cases of 2-dimension and 3dimension, respectively. Further from these results, it is easily understood that the total sales numbers of apparel change proportional to tn/2 in general dimension n. From the data obtained by actual company in Japan, it becomes clear that total sales numbers of men's shirt and women's blouse change proportional to t1.5 and t.
ノート
  • 井関 紗代, 北神 慎司
    2018 年 17 巻 5 号 p. 531-536
    発行日: 2018年
    公開日: 2018/12/26
    [早期公開] 公開日: 2018/11/13
    ジャーナル フリー
    Just imagining touching an object (i.e., haptic imagery) has been found to increase psychological ownership for an object. Our previous study revealed that purchase intention also increases as psychological ownership increases via perceived control promoted by haptic imagery. Then, how can the effect of haptic imagery on psychological ownership be promoted more? In this study, we investigated whether this effect could be promoted when effectance motivation is high. Effectance motivation is the motivation to feel competent to change the environment as a result of their activities and getting a sense of control. This study showed the following. Although only for objects that were actually touched previously, when effectance motivation is high, haptic imagery directly increased psychological ownership and this effect was fully mediated by perceived control. This study gives great suggestions in understanding consumer behavior when objects cannot be touched (e.g., online shopping).
  • 川口 めぐみ
    2018 年 17 巻 5 号 p. 557-560
    発行日: 2018年
    公開日: 2018/12/26
    [早期公開] 公開日: 2018/12/07
    ジャーナル フリー
    Different studies have examined impression formation regarding shapes. However, few studies have examined whether shapes directly evoke the emotions. An experiment was conducted to investigate characteristics of forms that evoke negative emotions among adolescents and young children. The results indicated that both adolescents and young children had negative emotions regarding circle-aggregate shapes. It might be possible that characteristics of circle-aggregate shapes include signals that developmentally and evolutionally facilitate approach-avoidance behaviors.
feedback
Top