Comics are one of the most popular narrative entertainment forms, with many titles offering varied types of stories. Although comic content itself has been analyzed, the way comic titles are enjoyed remains unrevealed and unexplored. We focus on comic shelves, where several comic titles are stored based on the Kansei of each user, serving as a data source for Kansei similarities among comic titles. We developed a platform to share digital shelves of comics with a certain theme with others: ComiNet. ComiNet displays bookshelves created by users, along with a network of comic titles based on their relationships stored in these bookshelves. Analysis of bookshelves obtained through a pilot study suggested that 1) others’ bookshelves encouraged users to share their Kansei with others, 2) bookshelf themes reflected diverse Kansei beyond standardized information such as genre or author, and 3) how comics are implicated was varied depending on the bookshelf theme.

We analyzed the relationship between speaker’s personalities and utterance features using dialogue corpus with personality information. We considered two kinds of personality: personality traits obtained from personality tests, and personality impressions inferred by other people from dialogue logs. We considered occurrence frequency of dialogue acts and specific vocabulary categories as utterance features. The results of the correlation analysis showed that while there were weak correlations between personality traits and utterance features in dialogue, many utterance features had moderate correlations with personality impressions. To realize a dialogue system that gives users a specific personality impression, we constructed a response generation model from pre-trained language model T5 by fine-tuning using this dialogue corpus and personality impression data and conducted an experiment on the reproducibility of personality. The experimental results showed that our response generation model can generate responses that reflect the high or low personality impression specified at the time of generation.

The paper proposes a method for detecting quality attributes that affect customer satisfaction from questionnaire data. By applying SHAP to the analysis of customer satisfaction survey data, we made it possible to semi-automatically classify quality elements based on the Kano model. In the proposed method, a Kano model is constructed through knowledge extraction from a learning model using questionnaire data as training data. As a result, existing customer satisfaction survey data can be utilized without investigating the degree of dependency or non-dependence of each quality characteristic. The proposed method offers many benefits to analysts, such as improved efficiency through effective use of data and assistance in extracting knowledge from data. In this study, the proposed method was applied to large-scale questionnaire data from the Oricon Customer Satisfaction Survey to demonstrate its effectiveness.

Film containers, which are commonly used in daily life, require not only sealing performance and protection of their contents but also ease of opening. This study investigated factors related to openability from the perspective of Kansei Engineering. Since opening a film container involves coordination between one hand that pinches the film and the other that supports the container, grip force was measured for the pinching hand and electromyography was recorded for the supporting hand. The results showed that grip force was regulated according to differences in the peeling strength of four types of film containers, and it is considered that moderate peeling strength provided better openability. Moreover, the coordination between both hands was found to play an essential role in opening film containers easily. These findings suggest that improving openability requires not only modification of film properties but also human-centered research.

This study aimed to present a scientific approach for determining signboard color combinations that convey a desired brand image while harmonizing with urban landscapes. Three impression rating experiments using the semantic differential method were conducted. The first experiment examined the relationship between two-color combinations and perceived impressions using simple geometric shapes, providing general guidelines. The second applied these findings to actual signboard configurations and identified combinations that achieved the target impression of “Aggressive Yet Elegant.” The third verified visual harmony and conspicuity of the selected configuration within real landscapes. Results indicated that a black background with white letters and lime-green ornamental lines was perceived as powerful, stylish, and visually prominent while maintaining harmony with surroundings. These findings offer practical insights for color planning in signboards and other visual media.

The empirical research in this paper is conducted on investment trust transactions. This empirical research targets the wealthy and is conducted to determine, among existing customers of financial institutions, what types of products the truly wealthy will be interested in as their lifestyles change, based on past data and the lifestyles of other similar wealthy individuals. This is then analyzed using machine learning to identify potential wealthy customers who have not yet made a transaction and are likely to be interested in the products proposed by financial institutions, and the accuracy and effectiveness of the analysis results are confirmed.

In recent years, social isolation and loneliness have become serious societal issues, increasingly affecting even younger generations. One contributing factor is the fragmentation of communities caused by life events such as relocation or transition to new life stages. To enhance the sense of connectedness among individuals in fragmented communities, we propose an interface design for a location-sharing system that visualizes behavioral commonality and aggregates users’ locations on a small-scale virtual map, aiming to foster awareness of others’ presence and activities. We verified the impact of the two factors in this method on users’ impressions of the provided information from three perspectives: sense of connectedness, sense of unity, and awareness. The results showed that the behavioral commonality was the primary factor contributing to all three aspects. Furthermore, the combination of the behavioral commonality with the aggregation of location information was effective in enhancing a sense of connectedness and unity.

This study investigates how distinct fabric softener fragrances influence emotional responses and perceptions during the laundry process. Three types of fabric softeners were used, each with a unique fragrance profile: Softener T (powdery/fruity/floral), Softener U (fruity/floral), and Softener V (fruity/woody/floral). A total of 34 female participants were divided into four groups, categorized by the duration of product use and the number and variety of softeners provided: Group P (4 weeks): All three softeners (T, U, and V); Group Q (1 week): All three softeners (T, U, and V); Group R (4 weeks): One softener of the participant’s choice; and Group S (1 week): One softener of the participant’s choice. Participants used the assigned softeners in their homes. Semistructured interviews were conducted to explore the emotional and perceptual impacts of softener use during laundry. The study examined how the number and variety of softeners and duration of use affected the participants’ experiences. Responses were analyzed using the KJ method, resulting in a two-part model of the laundry experience: (i) experiencing fragrances fostered positive impressions of doing laundry, including feelings of surprise and anticipation; (ii) experiencing multiple fragrances prompted reflection on desired emotional outcomes during laundry.