Works in detail

[Abstract] 

With advancements in artificial intelligence (AI), explainable AI (XAI) has emerged as a promising tool for enhancing the explainability of complex machine learning models. However, the explanations generated by an XAI may lead to cognitive biases among human users. To address this problem, this study aims to investigate how to mitigate users’ cognitive biases based on their individual characteristics. In the literature review, we found two factors that can be helpful in remedying biases: 1) debiasing strategies that have been reported to potentially reduce biases in users’ decision-making via additional information or change in information delivery, and 2) explanation modality types. To examine these factors’ effects, we conducted an experiment with a 4 (debiasing strategy) X 3 (explanation type) between-subject design. In the experiment, participants were exposed to an explainable interface that provides an AI’s outcomes with explanatory information, and their behavioral and attitudinal responses were collected. Specifically, we statistically examined the effects of textual and visual explanations on users’ trust and confirmation bias toward AI systems, considering the moderating effects of debiasing methods and watching time. The results demonstrated that textual explanations lead to higher trust in XAI systems compared to visual explanations. Moreover, we found that textual explanations are particularly beneficial for quick decision-makers to evaluate the outputs of AI systems. Next, the results indicated that the cognitive bias can be effectively mitigated by providing users with a priori information. These findings have theoretical and practical implications for designing AI-based decision support systems that can generate more trustworthy and equitable explanations.

[Abstract] 

Majority opinions are often observed in online environments. Previous studies have demonstrated that majority opinions are constructed because people with minority opinions have fear of isolation, which forces them to be silent. However, we often observe that online users with different minority opinions fight each other, even though there are only a few. To explain this phenomenon, we developed a new theoretical model and examined it through the analysis of Reddit data. The results show that users in small communities expressed relatively homogeneous and less negative opinions, and constructed majority opinions that were not as strong as those in large communities. Contrarily, users in large communities showed relatively heterogeneous and more negative opinions, and built majority opinions more strongly than those in small communities. This implies that the model can properly explain why and how majority-opinion users are silent and why minority-opinion users are loud.

[Abstract] 

Weak signals are rarely identified in the initial stage of growth and appear significant over time, unlike strong signals clearly observed in past trends. Weak signals are important cues that need to be analyzed to rapidly and accurately predict changes in the uncertain future. Researchers have developed various methods for identifying cues that can be significantly used for prediction. However, in many cases, they heavily depend on the opinions of experts or are applicable only to weak signals in specific fields. This study proposes a weak signal detection method that extracts weak signals by selecting significant keywords from literature database and grouping relevant keywords. Furthermore, this study presents a weak signal prediction method for predicting the growth of specific weak signals by investigating and learning the growth of the extracted weak signals over 10 years. To verify the proposed method, we extracted weak signals for 10 years (2001–2010) from SCOPUS publication data from 1996 to 2009 and applied machine learning using a graph convolutional network (GCN) model with the growth data of the extracted weak signals. The results showed that the proposed methods can effectively detect and predict weak signals.

[Abstract] 

Social networking services have been placed where people share opinions and information about various topics. These services allow users to express their opinions in direct (e.g., writing a comment or reply) and indirect ways (e.g., clicking a Like button). Based on commending, replying, and liking activities, users construct majority opinions in online environments. Previous studies examined perceptual and behavioral characteristics in the circumstance of majority opinions but only few of them provided how they differ depending on content types. Based on three different types of YouTube channels (news, documentary, and comedy), this study addresses how statistical properties of user opinions and majority opinions in online environments are presented differently depending on types of content. Based on the results of statistical analyses, we provide detailed properties of user activities in three types of YouTube channels and discuss several theoretical and practical implications.

[Abstract] 

A company’s benchmarking strategy is significantly determined by how it measures technological similarity. Researchers have measured the technological similarity between companies using a vector composed of the classification codes of patents that each company owns. However, patent classification code-based company comparison methods do not consider the text in patents and thus may not find similar companies accurately. To solve this problem, this study suggests a patent embedding-based company comparison method. The suggested method uses a text embedding model to vectorize the text in patents and calculates technological similarity based on the embedding vector. We examine the effectiveness of the suggested method by comparing it with the conventional patent classification code-based method. From the validation results for 11,227 Korean companies listed in the Korea Data Analysis, Retrieval, and Transfer system (DART), we find that the suggested method effectively retrieves technologically similar companies.

[Abstract] 

Bitcoin prices have fluctuated greatly, and news media have warned investors about a possible price bubble, arguing that the fluctuation arises mainly from people's blind pursuit of short-term trends. Despite these increasing concerns, however, only a few studies have addressed them. This article examines the problem using agent-based modeling. In our model, agents are designed to interact with one another in two ways: Price and social interactions. In their price interactions, agents adopt one of three strategies (fundamentalist, momentum trading, and contrarian trading) in investing at each time and adopt a strategy to maximize their expected benefits. In their social interactions, uninvolved agents become involved at various times based on their network properties (word-of-mouth effect). To examine the distinctive properties of Bitcoin from a comparative perspective, two representative currencies (Euro and Turkish lira) and two financial assets (Nasdaq and Nasdaq leverage index) are used in the agent-based model. The results show that the fraction of fundamentalists and price volatility have mutual Granger causal relationships overall. Also, no significant differences are found in the parameters of social interaction. These results are contrary to commonly held beliefs that Bitcoin prices are merely a result of blind pursuit and herding behavior.

[Abstract] 

Bitcoin prices have fluctuated greatly, and news media have warned investors about a possible price bubble, arguing that the fluctuation arises mainly from people's blind pursuit of short-term trends. Despite these increasing concerns, however, only a few studies have addressed them. This article examines the problem using agent-based modeling. In our model, agents are designed to interact with one another in two ways: Price and social interactions. In their price interactions, agents adopt one of three strategies (fundamentalist, momentum trading, and contrarian trading) in investing at each time and adopt a strategy to maximize their expected benefits. In their social interactions, uninvolved agents become involved at various times based on their network properties (word-of-mouth effect). To examine the distinctive properties of Bitcoin from a comparative perspective, two representative currencies (Euro and Turkish lira) and two financial assets (Nasdaq and Nasdaq leverage index) are used in the agent-based model. The results show that the fraction of fundamentalists and price volatility have mutual Granger causal relationships overall. Also, no significant differences are found in the parameters of social interaction. These results are contrary to commonly held beliefs that Bitcoin prices are merely a result of blind pursuit and herding behavior.

[Abstract] 

Contrary to the traditional concept of artificial intelligence, explainable artificial intelligence (XAI) aims to provide explanations for the prediction results and make users perceive the system as being reliable. However, despite its importance, only a few studies have investigated how the explanations of an XAI system should be designed. This study investigates how people attribute the perceived ability of XAI systems based on perceived attributional qualities and how the power status of the XAI and anthropomorphism affect the attribution process. In a laboratory experiment, participants (N = 500) read a scenarios of using an XAI system with either lower or higher power status and reported their perceptions of the system. Results indicated that an XAI system with a higher power status caused users to perceive the outputs of the XAI system to be more controllable by intention, and higher perceived stability and uncontrollability resulted in greater confidence in the system’s ability. The effect of perceived controllability on perceived ability was moderated by the extent to which participants anthropomorphised the system. Several design implications for XAI systems are suggested based on our findings.

[Abstract] 

The rapid change in technology makes it challenging to forecast the future of jobs. Previous studies have analyzed economics and employment data or employed expert-based methods to forecast the future of jobs, but these approaches were not able to reflect the latest technology trends in an objective way. To overcome the issue, this study matches jobs with patents and forecasts the future of jobs based on changes in the number of patents with time. A word embedding model is trained by patent classification code and job description data and used to find similar patent classification codes of jobs. For an illustration purpose, we identify information technology-related jobs listed in O*NET and discover similar patent classification codes of the jobs. Based on the change in the number of patents, we find promising jobs presenting high technical demands. Several implications of our approach are also discussed.

[Abstract] 

For user experience studies, the affordance theory has been used to describe a user’s intuitive perception and action. This theory, however, has often faced problems in application due to the different viewpoints of the ecological psychology and other fields of application study. To address this issue, we adopt two strategies in this study. First, we review the existing explications of the affordance theory in ecological psychology and rectify issues that have hindered the use of concepts in the affordance theory for user experience analysis. In addressing these issues, we suggest revised formal expressions and propose a new typological system. Second, by organizing the revised formal expressions and the new typology of the affordance into a heterophenomenological frame, we suggest a research framework for user experience analysis. We present an application example of the framework for demonstration purposes. We expect that the suggested framework will enable better descriptions of various phenomena occurring in the physical, social, and self dimensions and designs of products and services.

[Abstract] 

Despite technological advances, trust still remains as a major issue facing autonomous vehicles. Existing studies have reported that explanations of the status of automation systems can be an effective strategy to increase trust, but these effects can differ depending on the forms of explanations and autonomous driving situations. To address this issue, this study examines the effects of explanation types and perceived risk on trust in autonomous vehicles. Three types of explanations (i.e., no, simple, and attributional explanations) are designed based on attribution theory. Additionally, four autonomous driving situations with different levels of risk are designed based on a simulator program. Results show that explanation type significantly affects trust in autonomous vehicles, and the perceived risk of driving situations significantly moderates the effect of the explanation type. At a high level of perceived risk, attributional explanations and no explanations lead to the lowest and highest values in trust, respectively. However, at a low level of perceived risk, these effects reverse.

[Abstract] 

This study examines how repetitive use in a short time period influences perceived usability, aesthetics, and user satisfaction. A mixed design was used, considering usability, aesthetics, and exposure time. Sixty-four users were tested in a controlled experimental setting. According to the results, the negative effect of low usability on perceived usability and user satisfaction decreased as familiarity with the tasks increased. Also, the negative effect of low aesthetics on perceived usability weakened with repetitive use. These results imply that repetitive use can be an important matter for products with bad designs and promoting repetitive use in a short time period can be an effective strategy to reduce negative user perceptions.

[Abstract] 

Autonomous vehicles are expected to be commercialized within a few years, and researchers have investigated various factors that influence their adoption. However, only a few studies have considered comparative and psychological perspectives that can affect user-vehicle relationships. Focusing on this limitation, this study investigates influential factors on the use of autonomous vehicles in terms of a technology acceptance model (which considers perceived ease of use, perceived usefulness, and intention to use) and factors for autonomous vehicle use (e.g., perceived risk, relative advantage, self-efficacy, and psychological ownership (i.e., feeling of ownership)). Our results show that self-efficacy positively affects the perceived ease of use and intention to use, while the relative advantage affects perceived usefulness. Psychological ownership affects the intention to use but not the perceived usefulness. This implies that encouraging a consumer to form a psychological bond (i.e., psychological ownership) with an autonomous vehicle may be an effective strategy for promoting the use of autonomous vehicles.

[Abstract] 

People who are suspected to suffer mental disorders often explore online communities to gather medical information. Such medical information benefits these people by facilitating self-diagnosis and social support for the mental disorders. At the same time, however, misinformation can aggravate mental disorders and worsen psychological status. Focusing on two representative mental illnesses, bipolar and depressive disorders, this study analyzed how users shared their experiences with illness and provided advice. Postings for bipolar and depressive disorders were gathered from subreddit communities and used for semantic network analysis. Results showed that users in both communities described sleep disorder episodes and financial problems with negative emotional expressions. Users in the bipolar disorder community showed more interest in the topic of medication, whereas users in the depressive disorder community were more interested in suicide issues. We discuss how these properties in the subreddit communities can be applied to understand user experiences of bipolar and depressive disorders.

[Abstract] 

Majority opinions are often observed in the process of social interaction in online communities, but few studies have addressed this issue with empirical data. To identify an appropriate theoretical lens for explaining majority opinions in online environments, this study investigates the skewness statistic, which indicates how many “Likes” are skewed to major comments on a Facebook post; 3489 posts are gathered from the New York Times Facebook page for 100 days. Results show that time is not an influential factor for skewness increase, but the number of comments has a logarithmic relation to skewness increase. Regression models and Chow tests show that this relationship differs depending on topic contents, but majority opinions are significant in overall. These results suggest that the bandwagon effect due to social affordance can be a suitable mechanism for explaining majority opinion formation in an online environment and that majority opinions in online communities can be misperceived due to overestimation.

[Abstract] 

Various interfaces have been suggested for mobile devices, including touch gesture and embedded sensor-based interfaces. However, if a user's task requires a thorough look, these interfaces hinder sight of view in display and thus can be inappropriate for the tasks. This problem is more important especially in the mobile computer aided design (CAD) context, which performs visually demanding tasks on the limited screen. To address this point, this study suggests a mobile interface utilizing the rear camera of the device and describes the function mapping for CAD. The suggested interface detects finger attachment and movement direction. The number of colors in camera vision and local binary pattern are used as features, and a support vector machine (SVM) is used for feature classification. A prototype application is designed for validation, and appropriate SVM models are selected through benchmarking tests. The validation results show that the suggested interface can perform with high accuracy and low computational resources.

[Abstract] 

Since the 2010s, various companies have begun to manufacture wearable smartwatch devices, but the current sales of these products are not impressive. This study investigates how the limitations of the smartwatch are related to perceptual discomforts. Theoretically, this study evaluates the claim that the discomfort that users appear to have with the smartwatch stem from failed remediation. Users perceive the smartwatch more as a set of functional sensors rather than a watch or smartphone. Specifically, from the remediation perspective, the authors asked how users perceive the functions of the smartwatch. This study used dynamic topic modeling for topics on the smartwatch on Reddit. This study reports that the smartwatch has failed to provide a proper way to use the remediated content that it provides. Suggestions for future studies are addressed.

[Abstract] 

Purpose: The purpose of this paper is to investigate how we can understand social media interactions better by explicating the process of social capital formation on Facebook from a reciprocity perspective.
Design/methodology/approach: This study observed users who got tagged on Facebook by his/her friends and how s/he responded to that tagging activity. In total, 4,666 posts and 418,580 comments from The New York Times Facebook page were collected for the observation.
Findings: A majority (77.87 percent) of users who were tagged by their friends showed reactions to their tagging. In detail, 33.63, 44.20, and 0.04 percent of users responded by comments, “Likes”, and “Shares”, respectively. In total, 90.11 percent of the comments and 98.58 percent of the “Likes” were expressed on a comment or sub-comment, and only 9.89 percent of the comments and 1.42 percent of the “Likes” were expressed on a post. This indicates that a high percentage of users respond to their tagging notification, and they prefer dialogic responses to non-dialogic responses.
Originality/value: Previous studies have focused on photo tagging activity in social media, but user tagging activity had not been studied enough. This study examines the effects of Facebook tagging activity from a reciprocal perspective.

[Abstract] 

Recommendation systems are becoming important with the increased availability of online services. A typical approach used in recommendations is collaborative filtering. However, because it largely relies on external relations, such as items-to-items or users-to-users, problems occur when the relations are biased or insufficient. Focusing on that limitation, we here suggest a new method, item-network-based collaborative filtering, which recommends items through four steps. First, the system constructs item networks based on users’ item usage history and calculates three types of centrality: betweenness, closeness, and degree. Next, the system secures significant items based on the betweenness centrality of the items in each user's item network. Then, by using the closeness and degree centrality of the secured items, the algorithm predicts preference scores for items and their rank orders from each user's perspective. In the last step, the system organizes a recommendation list based on the predicted scores. To evaluate the performance of our system, we applied it to a sample dataset of 196 Last.fm users’ listening history and compared the results with those from existing collaborative filtering methods. The results showed that the suggested method performed better than the basic item-based and user-based collaborative filtering methods in terms of Accuracy, Recall, and F1 scores for top-k recommendations. This indicates that an individual user's item relations can be utilized to remedy the problems occurring when the external relations are biased or insufficient.

[Abstract] 

This study proposes a new approach to developing a user behavior model to explain how a user finds the optimal use. This is achieved by considering user concerns, task significances, affordances, and emotional responses as the interaction components and by exploring behavior sequences for a goal in using a product at the first time. The tasks in the same group at each level in the user concern structure are therefore in a competing relationship in going up to a higher task. The task tree with the significances and the affordance probabilities can be analyzed. The order of a user’s exploring behavior sequences can be determined through comparisons of the expected significances, which can be obtained by the modified Subjective Expected Utility Theory. A user’s emotional responses for the tasks which a behavior sequence is composed of can be calculated by the modified Decision Affect Theory. Here, the emotional response refers to a user’s internal reactions for the degree of which a product’s affordance features can meet his or her mental model in use. The average emotional response for a behavior sequence can be a user’s decisional factor for the optimal use method in using a product with a goal. Also, the design problems of a product can be checked from users’ point of view, and the emotional losses/changes by usage failures can be discussed. For an illustrative purpose, the proposed model is applied to a numerical example with some assumptions.