When using online video platforms like YouTube and Netflix, users often adjust playback speed, skip forward, or rewind. In this study, we developed a mobile app to collect real-world viewing logs and satisfaction responses and conducted a field study with 25 participants. Our findings reveal that specific playback behaviors, such as scrubbing and backward skipping, are closely associated with user engagement patterns. Importantly, we show that video abandonment does not always signal dissatisfaction. These insights provide valuable guidance for designing smarter recommendation algorithms and improving user experiences in online video streaming services.

As online fan communities continue to thrive, understanding how fans interact with artists and each other has become more important than ever, influencing both the vibrancy of the community and long-term engagement. In this study, we analyzed large-scale user data from Weverse, a global fandom platform, by applying k-means clustering and linguistic analysis to users' posting and commenting behaviors. Our analysis identified two distinct types of superfans: post-heavy users, who tend to post organized, goal-driven messages, and comment-heavy users, who engage more emotionally through casual language and frequent emoji use. These findings highlight opportunities for fan platforms to design features that better support diverse engagement styles and foster stronger community connections.

In online video streaming services such as YouTube, viewers occasionally encounter violent scenes featuring graphic blood. Completely removing these scenes would eliminate important contextual elements necessary for understanding the video, yet leaving them unchanged can make the viewing experience uncomfortable. In this study, we developed an AI model that selectively applies a blurring effect only to the blood, allowing users to watch the content more comfortably while still preserving its context. We compared viewers’ eye responses (specifically, whether blurring the blood affects blink rate or the extent of eye opening) under conditions with and without the model. This research represents a step forward in enhancing the overall viewing experience.

How many smartphone notifications do you receive in a day? Smartphone notifications can be overwhelming, disrupting work and causing fatigue. While some solutions use sensors or context to reduce interruptions, they often overlook what people personally need. We built a system with three modes—Immediate, While in Use, and On Demand—so you decide when and how to get notifications. In our two-week field study with 26 participants, it reduced notification volume, cut down on habitual phone-checking, and kept overall usage steady. Participants also reported lower addiction and fatigue, showing how customizing notifications helps you stay connected without feeling bombarded.

User-driven intervention tools such as self-tracking help users to self-regulate problematic smartphone usage. This paper proposes GoldenTime, a mobile app that promotes self-regulated usage behavior via system-driven proactive timeboxing and micro-financial incentives framed as gain or loss for behavioral reinforcement. We conducted a large-scale user study (n = 210) to explore how our proactive timeboxing and micro-financial incentives influence users’ smartphone usage behaviors.

Vehicle dashboard cameras or dashcams, among other smart vehicle technologies, are increasingly attracting interest across the globe. Furthermore, dashcam videos as objective witnesses are often shared to resolve various traffic incidents. In this work, we aim to understand cross-national differences in motives and privacy concerns of dashcam video-sharing, which are closely related to the factors that vary across countries, such as cultural values, traffic regulation, driving environments, and privacy perception.

What do you think of community safety? In this study, we designed a community-sourced patrolling campaign in which community members would schedule their patrol times and routes, perform bike-based patrolling with video capturing and share the information using their smartphones. We conducted a four-week field study (n=20) on a university campus to verify the campaign's feasibility and observe users' behavior. The results show key findings about users’ task scheduling, event capturing and reporting behaviors, factors affecting task selection and execution, and user motivation and engagement.

On computer-mediated communication, stickers or GIFs, though similar in appearance to emoji, have distinct characteristics because they often contain animation, various gestures, and multiple characters and objects. Their complexity and placement constraint may result in miscommunication. In this study, we aim to understand how people perceive emotion in stickers, as well as how miscommunication related to sticker occurs in actual chat contexts via online survey and interview.

In location-based social Q&A services, people ask a question with a high expectation that local residents who have local knowledge will answer the question. Besides, questions are classified by regions rather than topics, unlike existing topic-based Q&A services. We analyzed a 12-month period Q&A dataset from Naver KiN "Here," a popular location-based social Q&A mobile app in South Korea, in addition to a supplementary survey dataset obtained from 285 mobile users.

We often see various traffic violations on the road while driving. In this study, we consider community policing on the road with pervasive recording technologies. We developed a mobile app so that drivers can easily capture and report various threats to traffic safety to the police via mobile apps. Our two-week user study showed that the mobile app effectively supported community policing activities on the road.

What if your parent knows your bad driving habits? Bad driving behaviors are a major cause of traffic accidents, many of them resulting in fatalities. Our study intends to encourage safe driving habits by increasing drivers' self-awareness about their driving habits, as well as receiving supportive feedback on their driving behavior from a loved one as an intervention method. We built an Android prototype app to deliver feedback for bad driving maneuvers to both drivers themselves and to their corresponding supporters and conducted a field study evaluation.

Dashcams support the continuous recording of external views that provide evidence in case of unexpected traffic-related accidents and incidents. Recently, the sharing of dashcam videos has gained significant traction for accident investigation and entertainment purposes. Furthermore, there is a growing awareness that dashcam video sharing will significantly extend urban surveillance. Our work aims to identify the major motives and concerns behind the sharing of dashcam videos for urban surveillance.

Uber and Airbnb, two well-known sharing economy services, are facing conflicts with traditional taxi and hotel companies because these services have monetary benefits but are free from legacy regulations. However, non-monetary-based sharing services, represented by Couchsurfing, Inc., are free from such conflict and still successful. We investigated the distinctive user participation motivation of non-monetary-based sharing services versus monetary-based ones.

Our goal was to design a tool that can facilitate Q&A activities in offline presentations. We first identified several problems associated with current offline Q&A practices. We then developed SlideQA, an online slide-based Q&A tool, and explored its usability.

In location-based social Q&A, the questions related to a local community (e.g., local services and places) are typically answered by local residents. We wanted to deepen our understanding of the localness of knowledge sharing through investigating the topical and typological patterns related to the geographic characteristics, geographic locality of user activities, and motivations of local knowledge sharing. To this end, we analyzed a 12-month period Q&A dataset from Naver KiN "Here" and a supplementary survey dataset from 285 mobile users.