News events often generate fragmented coverage across a flood of articles, making it difficult for readers to discern what happened, what issues are at stake, and how different stakeholders are responding. While large language models (LLMs) are widely used to summarize articles, their use in synthesizing coherent digests from multiple sources about the same event remains underexplored. We present a study examining how different LLM prompting strategies—core issue, standpoints, and sub-events—shape readers’ perceptions of credibility, readability, and completeness. Based on a user evaluation (n=80) and follow-up interviews, we find that while standpoint digests improve perceived credibility, trust extends beyond structure. Sub-event digests support understanding of event development, and core-issue digests are often seen as vague. Readers also preferred timeline-based framing for ongoing events and concise, outcome-focused summaries for concluded ones. Consequently, we propose design principles emphasizing source transparency, verifiable evidence, visual scaffolding, and epistemic humility to foster trust.

Volunteered geographic information (VGI) platforms often suffer from redundant submissions, as contributors tend to create new reports rather than update existing ones—partly due to the effort required to locate and examine relevant prior entries. This study introduces Tell2Find, a large language model (LLM)-powered agent designed to support report retrieval and reuse during mobile crowdsourcing tasks. We conducted a study comparing Tell2Find with a traditional filter-based tool and a hybrid version combining both. Users with access to Tell2Find were more likely to examine existing reports before acting, resulting in more updates to prior reports and lower redundancy. Qualitatively, participants highlighted Tell2Find’s advantages in reducing the effort of matching reports, improving convenience in mobile contexts, and encouraging engagement with prior data due to more relevant suggestions. However, we also observed friction points: user trust around the system’s opaque matching logic and uncertainty about input phrasing occasionally led to disengagement with Tell2Find.

Prior work has examined how users judge their smartphone use, typically focusing on either usage duration or intention. How these two factors jointly shape such evaluations remains unclear. We conducted a two-week study with 104 participants, who reviewed their screenshots and provided labels of both usage intention and evaluation of time use. Across 73,000 sessions (6.1M screenshots), the relationship between duration and evaluation was initially linear but then bounded: positive evaluations declined and negative ones rose with longer phone use duration but both eventually stabilized, most often judged neutral. Trajectories varied by intention. Entertainment mirrored the overall trend; functional use continually lost positive evaluations, whereas information-seeking became increasingly positive during the first half hour before later declining; messaging-based connections slowly lost positive evaluations, while social media–based connections declined more quickly; finally, “no specific intention” unfolded in phases—from short positive use to regret-prone mid-length episodes to neutral long sessions.

Understanding how people are willing to express notification preferences is essential for designing personalized intelligent notification systems. Yet little is known about when, how, and under what conditions individuals choose to provide such input. We conducted semi-structured interviews with 33 participants, using design probes to examine the timing, methods, and concerns surrounding preference expression. Our findings make three contributions. First, we show that willingness to provide feedback depends not only on input ease and function but also on the justifiability of the moment, with requests embedded into notification-handling routines perceived as most natural. Second, we find that sustained engagement requires two forms of clarity: clarity in how to express one’s preferences and clarity in how the system interprets and applies that input. Third, we reveal expectations for notification systems to act as evolving partners that distinguish temporary and situational shifts from longer-term preference changes and support mutual learning over time.

Smartphone users often regret aspects of their phone use, especially social media use.However, pinpointing specific ways in which the design of an interface contributes to regrettable use can be challenging due to the complexity of social media app features and user intentions.We conducted a one-week study with 17 Android users, using a novel method where we passively collected screenshots every five seconds, which we analyzed via a multimodal large language model to understand participants' usage activity at a finegrained level.Triangulating this data with data from experience sampling, surveys, and interviews, we found that regret varies based on user intention, with non-intentional and social media use being especially regrettable.Regret also varies by social media activity; participants were most likely to regret viewing algorithmically recommended content and comments.Additionally, participants frequently deviated to browsing social media when their intention was direct communication, which slightly increased their regret.Our findings provide guidance to designers and policy-makers seeking to improve users' experience and autonomy.

To help users manage the overwhelming influx of smartphone notifications, this study explores how large language models (LLMs) can be leveraged to generate notification summaries. We developed an Android application that integrates ChatGPT to summarize notifications and conducted an in-the-wild deployment to examine how users guided the model. To further understand user expectations for LLM-generated summaries, we interviewed 20 participants following a week-long engagement with the app. Our findings reveal five main strategies that users employed in their prompts for generating summaries. Additionally, interviewees expected summaries to prioritize three types of notifications, preferred three levels of information disclosure influenced by content anticipation and perceived criticality, and used three different approaches to synthesizing notifications based on their interrelationships. Finally, interviewees envisioned notification summarization functioning like a virtual assistant, desiring capabilities beyond simple information condensation, including support for task and information management, revisiting archived content, and tracking activities for reflection.

In the realm of smart cities, sensor technologies play a pivotal role in monitoring urban facilities and environments, providing real-time, site-specific information to residents. However, discrepancies often arise in sensor data due to variances in granularity, abstraction, and scope, which can foster uncertainty regarding the actual conditions on-site. This study explores whether, under these circumstances, individuals prefer on-site mobile crowds for verification purposes or for the provision of supplementary contextual information to aid in decision-making. Conducting an online study with 100 participants from Taiwan, who engaged in a think-aloud process while utilizing smart city sensor data for decision-making, our findings indicate that participants more often (54%) preferred seeking verification over supplementary contextual information (46%). Both pre-existing expectations and the sense of task urgency affected participants' choices between verification and supplementary contextual information. However, we found that the driving factor for seeking supplementary contextual information was not sensor data deviating from pre-existing expectations, but rather the absence of such pre-existing expectations. Our qualitative data also uncovered five primary motivations and four factors influencing the choice of crowdsourced information. Overall, these findings contribute to our understanding of how people leverage on-site mobile crowds to supplement sensor data in the context of smart cities.

This three-phase study explores the experiential background of contributors to platforms that provide crowdsourced location-related information. Initially, we utilized interviews to understand users’ expectations for location-related information and the contributors’ experiential background they believe would enhance this information’s utility. We then deployed a survey to identify the top eight sought-after location-information types and their perceived characteristics. Then the concluding online scenario-based study provided quantitative evidence about the interrelationships of eight types of location-related information, ten crucial quality attributes, and aspects of the contributors’ experiential background believed to enhance the utility of the descriptions they provide. Notably, although certain experiential background aspects were deemed universally advantageous across all information types, unique connections were identified among specific information types and distinct experiential background aspects seen as augmenting the contributor’s descriptions’ utility. These insights underline the importance of location-based crowdsourcing platforms incorporating contributors’ experiential background when assigning tasks.

In this study, we developed predictive models to address the problem of chatbot abandonment, a problem that can result in losing business opportunities. Specifically, we target on a conversation log dataset of a banking chatbot involving 1,373 users and hand-crafted features. By leveraging a pre-trained BERT model on the textural features and the hand-crafted features, the model achieved an F1-score of 0.89 in predicting discontinued conversation and 0.80 in predicting abandonment. Our findings indicate that textual features help capture more abandonment, while hand-crafted features improve detection precision. Our analysis with SHAP and LIME revealed that user typing, the chatbot expressing inability of recognizing intent, and the chatbot asking what users want to do during an ongoing conversation are top signals of user abandoning the chatbot. These findings suggest that chatbot designers should consider providing pre-set options or constraints for user inputs and presenting possible intents to the user and avoid expressing inability, incompetence, or ignoring the users’ current attempt.

Research indicates that smartphone users often speculate about notifications upon sensing their arrival, aiding their decision to attend to them. This speculation, however, relies on the presence of sufficient clues to associate with the notification, which are not always available. To address this challenge, through an experience sampling study, we investigated the effectiveness of delivering user-assigned alerts in influencing users' speculation accuracy, attendance effectiveness, and perceived disturbance. Our findings suggest that while user-assigned alerts enhanced the accuracy of speculation and improved participants' decisions to attend to notifications, the increased notification awareness sometimes led participants to view their decision to ignore notifications as less favorable. Moreover, we found that sporadic alert delivery disrupted the association between the alert and the notification, leading to no reduction in perceived disturbance nor improvement in speculation accuracy. In assigning alerts to notifications, participants considered five strategies: familiarity, distinctiveness, disturbance, emotional resonance, and dimension representation.

In this exploratory experience sampling method (ESM) research, we examined the perceptions of 74 smartphone users regarding the opportuneness of moments for proceeding through a four-stage notification-response process: the phone generating an alert (Alert), the user roughly glancing at the notification (Glance), engaging with it (Engage), and acting on it (Act). We investigated how the moments perceived as opportune for each of the four stages related to users' self-reported values of 20 contextual factors, and how these factors influenced users' perceived opportuneness of the moments for each stage. Our results reveal that Alert and Glance stages were perceived as more distinct, with Alert being influenced by social-environmental related factors and Glance characterized by a lower threshold for what constitutes an opportune moment. The final two stages - Engage and Act - were the most similar to each other. The findings also indicated how the influence of contextual factors on perceived opportuneness of the moments varied across factors, notification types, stages, and how such variation was manifested in the likelihood, valence, and magnitude of their overall influence.

Notifications are commonly considered a distraction when they arrive during a task, and consequently, prior research has consistently sought effective ways of deferring their arrival until task transitions. However, many smartphone users still interact with notifications during tasks. In our qualitative study combining diary study and semi-structured interviews, we examined 34 research participants’ motivations for interacting with smartphone notifications at different times, including during tasks. Our findings resulted in a human-notification interaction framework comprised of 12 unique motivations frequently associated with three activity timings for interacting with notifications, including before-task, during-task, and after-task. Notably, participants frequently perceived interaction with notifications as a tool for improving task performance, making the most of their time, and promoting personal well-being, rather than only as a distraction. The before-the-task timing, in particular, has received little attention in previous research and deserves more attention as it was related to specific user motivations for notification interaction.

Time-killing on smartphones has become a pervasive activity, and could be opportune for delivering content to their users. This research is believed to be the first attempt at time-killing detection, which leverages the fusion of phone-sensor and screenshot data. We collected nearly one million user-annotated screenshots from 36 Android users. Using this dataset, we built a deep-learning fusion model, which achieved a precision of 0.83 and an AUROC of 0.72. We further employed a two-stage clustering approach to separate users into four groups according to the patterns of their phone-usage behaviors, and then built a fusion model for each group. The performance of the four models, though diverse, yielded better average precision of 0.87 and AUROC of 0.76, and was superior to that of the general/unified model shared among all users. We investigated and discussed the features of the four time-killing behavior clusters that explain why the models’ performance differ.

The proliferation of influencers on social-media platforms has drawn considerable research attention, particularly in the field of marketing. Nevertheless, there is limited understanding among HCI and communication researchers of what leads these social-media influencers’ (SMIs’) audiences to favor and choose their content over traditional media. To fill this gap, we conducted semi-structured interviews with 45 SMI audience members. Our findings revealed a total of eight categories of SMIs’ appeals, i.e., factors that made the interviewees favor their content over traditional media. These appeals can further be grouped into three categories: content, presentation, and closeness. In particular, we identified the key role of SMIs’ perceived high autonomy and independence, which led both their content and their presentation styles to be seen as distinct from and more appealing than traditional media. Likewise, four closeness appeals made our participants feel emotionally attached to SMIs, resulting in sustained engagement.

Mobile users commonly multitask during travel, but doing so on public transit can be challenging due to the dynamic nature of the environment as well as long-standing lack of infrastructural support. Nevertheless, HCI scholars and practitioners have devoted relatively little attention to developing technology for enhancing travel multitasking. To facilitate such development, we sought to understand travel multitaskers’ practices and challenges while on public transit, and to that end, conducted a multi-methods study that involved shadowing and interviewing 30 of them. We identified four travel-multitasking patterns, characterized by distinct motives that affected these travelers’ multitasking practices, receptivity to environmental stimuli, and task persistence. The two main challenges they encountered during travel multitasking resulted from mutual interference from their tasks and from the dynamic nature of transit environments. Based on these findings, design recommendations for public-transit agencies and mobile services are also provided.

People nowadays can use multiple devices to interact with notifications, whether via noticing, glancing, reading, or acting upon them. Prior research has focused on actual usage or on device preferences. However, users’ ideal experience of cross-device notification-interaction might differ from their current practices (due to situational limitations) and/or across the four notification-interaction stages. We therefore conducted an experience-sampling method study with multi-device users to investigate these gaps and the influence of device context. Our results reveal that nearly half of the time, the non-phone devices the participants had ranked as their top preferences for notification-interaction were not actually used, due to the devices’ context. Beyond device context, the participants’ choices of devices for notification-interaction were heavily determined by 1) their preferences that particular notification-interaction stages to take place (or not) on particular devices; and 2) the device on which they had undertaken the former stage.

Turn-by-turn navigation guidance is suggested to impair users’ independent wayfinding in the physical world. However, whether this impairment issue also exists in a virtual environment is underexplored. We compare map-based and live view-based turn-by-turn navigation aids with two additional navigation aids, reference-based and orientation-based, designed to provide directional knowledge. The results of our within-subjects experiment indicate that turn-by-turn navigation aids performed worse in building spatial awareness than reference-based guidance in virtual reality. In unaided wayfinding, reference-based guidance helped users navigate most efficiently. Unexpectedly, orientation-based guidance yielded poor navigation performance, similar to the turn-by-turn navigation aids. We found that the key to skill impairment is navigators’ tendency to rely on automated instructions. This suggests that turn-by-turn navigation aids need not be avoided, but rather that caution should be exercised to avoid the tendency of mindless instruction-following. Our qualitative findings also suggest the crucial roles of the navigation context and the suitability of the navigation aid for the context.

Customers learn about restaurants in various ways, and integrating this disparate information could give them access to a greater diversity of perspectives. Conflicting opinions between restaurant-review platforms are inevitable. However, such conflicts’ influences on users’ perceptions remain unclear, especially when the opinion of a user’s preferred platform conflicts with the majority of others. This study’s experiment with a sample of 304 users found that, when such situations occurred, the preferred platform’s influence differed depending on whether the user was shown a sequence of whole-platform aggregations vs. a sequence of individual reviews drawn from multiple platforms. That is, the participants accepted the majority view most of the time, but when looking at aggregated lists, if their preferred platform expressed a minority positive opinion based on a high quantity of reviews, that minority opinion could prevail over the majority one. Between-platform conflicts were also found to have a greater impact on user reactions than within-platform ones did.

News notifications on smartphones provide a convenient way to stay informed, but their delivery timing can influence user engagement. Despite this, research on the impact of notification timing on reading behavior remains limited. Therefore, we developed NewsMoment, a news aggregation app that monitors user reading patterns and sends news notifications. Our experience sampling study with 46 NewsMoment users revealed four distinct reading modes: typical, comprehensive, scanning, and unengaged. Deep reading, encompassing typical and comprehensive modes, more often occurred during self-initiated browsing rather than through pushed news. Interestingly, shallow reading modes - unengaged and scanning - showed varying prevalence, associated triggers, and engagement, despite their similarities. Importantly, unengaged reading persisted regardless of users' perceived moment opportuneness, whereas scanning reading was more common during inopportune moments. These findings suggest that identifying opportune moments for news reading may primarily reduce scanning reading, without substantially impacting unengaged reading.

Virtual reality (VR) has increasingly been used in many areas, and the need to deliver notifications in VR is also expected to increase accordingly. However, untimely interruptions could largely impact the experience in VR. Identifying opportune times to deliver notifications to users allows for notifications to be scheduled in a way that minimizes disruption. We conducted a study to investigate the use of sensor data available on an off-the-shelf VR device and additional contextual information, including current activity and engagement of users, to predict opportune moments for sending notifications using deep learning models. Our analysis shows that using mainly sensor features could achieve 72% recall, 71% precision and 0.86 area under receiver operating characteristic (AUROC); performance can be further improved to 81% recall, 82% precision, and 0.93 AUROC if information about activity and summarized user engagement is included.

Attention-management tools can restrict online communication, but may cause collateral damage to their users’ fulfillment of communication expectations. This paper explores the idea of integrating attention management into instant messaging (IM), by 1) disclosing restriction status via an online status indicator (OSI) to manage contacts’ expectations, and 2) imposing communication limits to reduce communication distraction. We used a speed-dating design method to allow 43 participants to rapidly compare 48 types of OSI restriction in various conversational contexts. We identified two “tug-of-wars” that take place when attention management is integrated into IM apps: one between fulfilling one’s contacts’ expectations and protecting one’s own attention, and the other, between protecting one’s privacy and asserting the justifiability of using communication restrictions. We also highlighted the participants’ desire to be diplomatic for sustaining their positive images and maintaining relational connectedness. Finally, we provide design recommendations for integrating attention management into IM apps.

We investigate instant-messaging (IM) users’ sense-making and practices around read-receipts: a feature of IM apps for supporting the awareness of turn-taking, i.e., whether a message recipient has read a message. Using a grounded-theory approach, we highlight the importance of five contextual factors – situational, relational, interactional, conversational, and personal – that shape the variety of IM users’ sense-making about read-receipts and strategies for utilizing them in different settings. This approach yields a 21-part typology comprising five types of senders’ speculation about why their messages with read-receipts have not been answered; eight types of recipients’ causes/reasons behind such non-response; and four types of senders’ and recipients’ subsequent strategies, respectively. Mismatches between senders’ speculations about un-responded-to read-receipted messages (URRMs) and recipients’ self-reported explanations are also discussed as sources of communicative friction. The findings reveal that, beyond indicating turn-taking, read-receipts have been leveraged as a strategic tool for various purposes in interpersonal relations.

The popularity of task-oriented chatbots is constantly growing, but smooth conversational progress with them remains profoundly challenging. In recent years, researchers have argued that chatbot systems should include guidance for users on how to converse with them. Nevertheless, empirical evidence about what to place in such guidance, and when to deliver it, has been lacking. Using a mixed-methods approach that integrates results from a between-subjects experiment and a reflection session, this paper compares the effectiveness of eight combinations of two guidance types (example-based and rule-based) at four guidance timings (service-onboarding, task-intro, after-failure, and upon-request), as measured by users’ task performance, improvement on subsequent tasks, and subjective experience. It establishes that each guidance type and timing has particular strengths and weaknesses, thus that each type/timing combination has a unique impact on performance metrics, learning outcomes, and user experience. On that basis, it presents guidance-design recommendations for future task-oriented chatbots.

“You eat with your eyes first.” Marcus Gavius Apicius, the insightful first-century Roman gourmand, stated. Although arranging foods in attractive ways can increase one’s appetite, creating an aesthetic food presentation is challenging. For instance, users have to cut ingredients into pieces of specific shapes and sizes, while imagining the overall appearance of their desired composition. To overcome such challenges, we introduce a system that assists users to arrange ingredients to present appealing patterns in meals. The system enables them to perform a process of trial-and-error in the simulation prior to creating a real food presentation. Due to machines’ high computing power, our automatic simulation provides users with a variety of food presentation results and inspires their creativity accordingly. It also computes the nutritional composition of each simulated food presentation so that both visual quality and health are considered simultaneously. Results demonstrate that the simulated food presentations are visually appealing and could be physically created. Participants who joined the user study also favored our food presentation simulation system.

Using a mixed-methods approach over six weeks, we studied 30 smartphone users’ task choices, task execution and effort devoted to two commercial mobile crowdsourcing platforms in the wild. We focused on the influence of activity contexts, characterized by breakpoint situations and activity attributes. In line with their stated preferences, the participants were more likely to proactively perform mobile crowdsourcing tasks during transitions between activities than during an ongoing activity and during long breaks, respectively. Their task choices were influenced by various activity attributes, and more impacted by their current and preceding activities than their upcoming ones. Two of our three target outcomes, task execution and task choice, were also influenced by individuals’ stress and energy levels. Our qualitative data provide further insights into participants’ decisions about which crowdsourcing tasks to perform and when; and our results’ implications for the design of future mobile crowdsourcing task-prompting mechanisms are also discussed.

Smartphone users do not deal with notifications strictly in the order they are displayed, but sometimes read them from the middle, suggesting a mismatch between current systems’ display order and users’ needs. We therefore used mixed methods to investigate 34 smartphone users’ desired notification display order and related it with users’ self-reported order of attendance. Classifying using these two orders as dimensions, we obtained seven types of notifications, which helped us not only highlight the distinct attributes but understand the implied roles of these seven types of notifications, as well as the implied meaning of display orders. This is especially manifested in our identification of three main mismatches between the two orders. Qualitative findings reveal several meanings that participants attached to particular positions when arranging notifications. We offer design implications for notification systems, including calling for two-dimensional notification layout to support the multi-purpose roles of smartphone notifications we identified.

Researchers have long attempted to estimate instant-messaging (IM) users’ attentiveness, responsiveness, and interruptibility. Yet, IM users’ self-presentation of their receptivity, and their perceptions of automated adjustment/revelation of their receptivity status (e.g., Facebook Messenger’s green dot that deems a user to be “active”), remain under-explored. We therefore told our 43 participants that our IM app, IMStatus, was capable of automatically estimating and adjusting their receptivity status to responsive, attentive, or interruptible based on their smartphone activity. These statuses were also presented to their IM contacts in three different styles. Over a two-week period, the participants rarely chose the status interruptible, and when they did, it was usually to indicate low availability. Textual presentation was usually chosen to express statuses precisely, especially at high and low extremes of receptivity; while graphical and numeric presentations were preferred when self-perceived receptivity levels were more ambiguous. Conflicts between recipients’ and senders’ perspectives are also discussed.

With the diversified sources satisfying users' needs, many online service platforms collect information from multiple sources in order to provide a set of useful information to the users. However, existing recommendation systems are mostly designed for single-source data, and thus fail to recommend multi-source review packages since the interplay between the reviews of different sources is not properly modeled. In fact, modeling the interplay between different sources is challenging because 1) two reviews may conflict with each other, 2) different users have different preferences on review sources, and 3) users' preferences to each source may change under different scenarios. To address these challenges, we propose Self-Attentive Recommendation for multi-source review Package (SARP), for predicting how useful the user feels to the package, while simultaneously reflecting how much the user is affected by each review. Specifically, SARP jointly considers the relationships of every user, purpose, and review source to learn better latent representations. A self-attention module is further used for integrating source representations and the review ratings, following a multi-layer perceptron (MLP) for the prediction tasks. Experimental results on the self-constructed dataset and public dataset demonstrate that the proposed model outperforms the state-of-the-art approaches.

Numerous systems for helping visually impaired people navigate in unfamiliar places have been proposed. However, few can detect and warn about moving obstacles, provide correct orientation in real time, or support navigation between indoor and outdoor spaces. Accordingly, this paper proposes V-Eye, which fulfills these needs by utilizing a novel global localization method (VB-GPS) and image-segmentation techniques to achieve better scene understanding with a single camera. Our experiments establish that the proposed system can reliably provide precise locations and orientation information (with a median error of approximately 0.27 m and 0.95°); detect unpredictable obstacles; and support navigating both within and between indoor and outdoor environments. The results of a user-experience study of V-eye further indicate that it helped the participants not only with navigation, but also improved their awareness of obstacles, enhanced their spatial awareness more generally, and led them to feel more secure and independent while walking.

AI-enabled smart-home agents that automate household routines are increasingly viable, but the design space of how and what such systems should communicate with their users remains underexplored. Through a user-enactment study, we identified various interpretations of and feelings toward such a system's confidence in its automated acts. That confidence and their own mental models influenced what and how the participants wanted the system to communicate, as well as how they would assess, diagnose, and subsequently improve it. Automated acts resulted from false predictions were not generally considered improper, provided that they were perceived as reasonable or potentially useful. The participants' improvement strategies were of four general types, all of which will be discussed. Factors affecting their preferred levels of involvement in automated acts and their interest in system confidence were also identified. We conclude by making practical design recommendations for the user-system communication design spaces of smart-home routine assistants.

Virtual reality (VR) platforms provide their users with immersive virtual environments, but disconnect them from real-world events. The increasing length of VR sessions can therefore be expected to boost users' needs to obtain information about external occurrences such as message arrival. Yet, how and when to present these real-world notifications to users engaged in VR activities remains underexplored. We conducted an experiment to investigate individuals' receptivity during four VR activities (Loading, 360 Video, Treasure Hunt, Rhythm Game) to message notifications delivered using three types of displays (head-mounted, controller, and movable panel). While higher engagement generally led to higher perceptions that notifications were ill-timed and/or disruptive, the suitability of notification displays to VR activities was influenced by the time-sensitiveness of VR content, overlapping use of modalities for delivering alerts, the display locations, and a requirement that the display be moved for notifications to be seen. Specific design suggestions are also provided.

Task-oriented chatbots are becoming popular alternatives for fulfilling users' needs, but few studies have investigated how users cope with conversational 'non-progress' (NP) in their daily lives. Accordingly, we analyzed a three-month conversation log between 1,685 users and a task-oriented banking chatbot. In this data, we observed 12 types of conversational NP; five types of content that was unexpected and challenging for the chatbot to recognize; and 10 types of coping strategies. Moreover, we identified specific relationships between NP types and strategies, as well as signs that users were about to abandon the chatbot, including 1) three consecutive incidences of NP, 2) consecutive use of message reformulation or switching subjects, and 3) using message reformulation as the final strategy. Based on these findings, we provide design recommendations for task-oriented chatbots, aimed at reducing NP, guiding users through such NP, and improving user experiences to reduce the cessation of chatbot use.

This study examines the characteristics of mobile instant-messaging users' relationships with their social contacts and the effects of both relationship and interruption context on four measures of receptivity: Attentiveness, Responsiveness, Interruptibility, and Opportuneness. Overall, interruption context overshadows relationship characteristics as predictors of all four of these facets of receptivity; this overshadowing was most acute for Interruptibility and Opportuneness, but existed for all factors. In addition, while Mobile Maintenance Expectation and Activity Engagement were negatively correlated with all receptivity measures, each such measure had its own set of predictors, highlighting the conceptual differences among the measures. Finally, delving more deeply into potential relationship effects, we found that a single, simple closeness question was as effective at predicting receptivity as the 12-item Unidimensional Relationship Closeness Scale.

Mobile devices such as smartphones and tablets have become our daily necessities. People spent a lot of time using their smartphones, especially on interactive applications such as social networks, web browsing, and games. To conserve power consumption of such applications becomes ultimately important. Prior works try to save power by balancing performance and workload. These approaches are not comprehensive because they do not take user experience (or Quality of Experience, QoE) into account. Overlooking this factor may turn out to subscribe an excessive performance that the user cannot even perceive, leading to a waste of power. Users' QoE is affected by many factors, including their contexts and the stage of the application. Estimating the QoE requirements for different situations is not only complicated but also crucial. In this paper, we attack this complex problem by first predicting the execution stages of the interactive applications and then applying a stage-based, context-aware, QoE-driven governor (SCQ) to adjust the CPU frequency. Experimental results show that our approach can save 23% more power than the default governor without affecting users' QoE.

Smartphone users' decisions about whether to attend to a notification after sensing it are under-researched. We therefore studied 33 Android users, and found that they speculated extensively about notifications' sources---i.e., which apps and which senders were responsible for them--- before attending to them. The participants' speculation about apps was both more common and more accurate than that about senders. They also were more likely to 1) perceive notifications as important, 2) attend to them, and 3) consider them beneficial if they speculated about them than if they did not or could not. Participants' speculations were based on the alert's inherent characteristics, context, and temporality. Inaccurate speculations were mainly caused by unclear signals, insufficient clues, and a multiplicity of possible sources. Ringer mode affected the accuracy of user speculation, but not its frequency or the frequency of attending to notifications.

The experience sampling method (ESM) is widely used for collecting in situ experiences in various domains. One known limitation, however, is its reliance on participants being receptive to ESM questionnaires at the sampled moments. At moments when participants cannot notice or respond to an ESM questionnaire, researchers cannot obtain a response. In this research, we explored the feasibility of inviting peers to provide information about participants in an ESM study. Results from a two-week experiment with a total of 27 participants and 82 peers showed that including peers' ESM responses increased ESM data quantity. Furthermore, the agreement between the peers' and the participants' responses could be maintained by asking peers' confidence. Even considering only data with high confidence could increase data quantity. Moreover, inviting peers had a positive impact on the participant's compliance to respond. These results suggest that using peer-ESM to obtain more in-situ data about participants is promising.

Current mobile instant messaging (IM) applications offer limited information on the availability status of IM users, particularly, their availability for reading and responding to IM messages. Research suggests a gap between what IM recipients want to disclose and what IM senders want to see to determine when to initiate a conversation. The advancement of IM users' receptivity prediction makes it possible to present IM users' predicted availability status. In this research, we conducted user enactment, a design approach for researchers to let participants experience and reflect on possible designs of future technologies, to explore designs of IM availability status in 72 IM conversation scenarios. We explore how IM users interpret different presentations of an uncertain IM status from both the senders' and recipients' perspectives, and what they need and will act upon these presentations.

History of conversations through instant messaging (IM) contains abundant information about the communication patterns of the dyad, including conversation partners' mutual responsiveness to messages. We have, however, not seen many examinations of using such information in modeling mobile users' responsiveness in IM communication. In this paper, we present an in-the-wild study, in which we leverage participants' IM messaging logs to build models predicting their general responsiveness. Our models based on data from 33 IM user achieved an accuracy of up to 71% (AUROC). In particular, we show that 90-day IM-communication patterns, in general, outperformed their 14-day equivalent in our prediction models, indicating better coherence between long-term IM patterns with their general communication experience.

Task-oriented chatbots are increasingly prevalent in our daily life. Research effort has been devoted to advancing our understanding of users' interaction with conversational agents, including conversation breakdowns. However, most research attempts were limited to obversions from a relatively short duration of user interaction with chatbots, where users were aware of being studied. In this study, we conducted a conversation analysis on a three-month conversation log of users conversing with a chatbot of a banking institution. The log consisted of 1,837 users' conversations with this chatbot with 19,449 message exchanges. From this analysis, we show that users more often failed to make a progress in a conversation when they requested information than when they provided information. Furthermore, we uncovered five kinds of intention gaps unexpected to the chatbot, and five major behaviors users adopted to cope with non-progress.

User-centric power management in smartphones aims to conserve power without affecting user's perceived quality of experience. Most existing works focus on periodically updated applications such as games and video players and use a fixed frame rate, measured in frame per second (FPS), as the metric to quantify the display quality of service (QoS). The idea is to adjust the CPU/GPU frequency just enough to maintain the frame rate at a user satisfactory level. However, when applied to aperiodically-updated interactive applications, e.g. Facebook or Instagram, that draw the frame buffer at a varying rate in response to user inputs, such a power management strategy becomes too conservative. Based on real user experiments, we observe that users can tolerate a certain percentage of frame drops when running aperiodically updated applications without affecting their perceived display quality. Hence, we introduce a new metric to characterize display quality of service, called the frame drawn ratio (FDR), and propose a new CPU/GPU frequency governor based on the FDR metric. The experiments by real users show that the proposed governor can conserve 17.2% power in average when compared to the default governor, while maintaining the same or even better QoE rating.

360° videos give viewers a spherical view and immersive experience of surroundings. However, one challenge of watching 360° videos is continuously focusing and re-focusing intended targets. To address this challenge, we developed two Focus Assistance techniques: Auto Pilot (directly bringing viewers to the target), and Visual Guidance (indicating the direction of the target). We conducted an experiment to measure viewers' video-watching experience and discomfort using these techniques and obtained their qualitative feedback. We showed that: 1) Focus Assistance improved ease of focus. 2) Focus Assistance techniques have specificity to video content. 3) Participants' preference of and experience with Focus Assistance depended not only on individual difference but also on their goal of watching the video. 4) Factors such as view-moving-distance, salience of the intended target and guidance, and language comprehension affected participants' video-watching experience. Based on these findings, we provide design implications for better 360° video focus assistance.

Watching a 360° sports video requires a viewer to continuously select a viewing angle, either through a sequence of mouse clicks or head movements. To relieve the viewer from this “360 piloting” task, we propose “deep 360 pilot” - a deep learning-based agent for piloting through 360° sports videos automatically. At each frame, the agent observes a panoramic image and has the knowledge of previously selected viewing angles. The task of the agent is to shift the current viewing angle (i.e. action) to the next preferred one (i.e., goal). We propose to directly learn an online policy of the agent from data. Specifically, we leverage a state-of-the-art object detector to propose a few candidate objects of interest (yellow boxes in Fig. 1). Then, a recurrent neural network is used to select the main object (green dash boxes in Fig. 1). Given the main object and previously selected viewing angles, our method regresses a shift in viewing angle to move to the next one. We use the policy gradient technique to jointly train our pipeline, by minimizing: (1) a regression loss measuring the distance between the selected and ground truth viewing angles, (2) a smoothness loss encouraging smooth transition in viewing angle, and (3) maximizing an expected reward offocusing on a foreground object. To evaluate our method, we built a new 360-Sports video dataset consisting offive sports domains. We trained domain-specific agents and achieved the best performance on viewing angle selection accuracy and users' preference compared to [53] and other baselines.

Collecting ground-truth annotations for contextual data is vital to context-aware system development. However, current research lacks a systematic analysis of different approaches to collecting such data. We present a field experiment comparing three approaches: Participatory, Context-Triggered In Situ, and Context-Triggered Post Hoc, which involved users in recording and annotating activity data in real-world settings. We compared the quantity and quality of collected data using each approach, as well as the participant experience. We found Context-Triggered approaches produced more recordings, whereas the Participatory approach produced a greater amount of data with higher completeness and precision. Moreover, while participants appreciated automated recording and reminders for convenience, they highly valued having control over what and when to record and annotate. We conclude that user burden and user control are key aspects to consider when collecting and annotating contextual data with participants, and suggest features for a future tool focused on these two aspects.

Smartphones are considered to be "always on, always connected" but mobile users are not always attentive and responsive to incoming communication. We present a mixed methods study investigating how mobile users use ringer modes for managing interruption by and awareness of incoming communication, and how these practices and locales affect their attentiveness and responsiveness. We show that mobile users have diverse ringer mode usage, but they switch ringer modes mainly for three purposes: avoiding interruption, preventing the phone from disrupting the environment, and noticing important notifications. In addition, without signals of notifications, users are less likely to immediately attend to notifications, but they are not less responsive to those they have attended. Finally, ringer mode switches, attentiveness, and responsiveness are all correlated with certain locales. We discuss implications from these findings, and suggest how future CMC tools and notification services take different purposes for using ringer modes and locales into consideration.

Transportation mode (TM) detection is one of the activity recognition tasks in ubiquitous computing. A number of previous studies have compared the performance of various classifiers for TM detection. However, the current study is the first work aiming to understand how TM detection performance is impacted by how the recorded location traces are segmented into data segments for training a classifier. In our preliminary experiments we examine three trace segmentation (TS) methods---Uniform Duration (UniDur), Uniform Number of Location Points (UniNP), and Uniform Distance (UniDis)---and compare their performance on detecting different transportation modes. The results indicate that while driving can be more accurately detected by using UniDis method, walking and bus can be more accurately detected by using UniDur method. This suggests that choosing a right TS method for training a TM classifier is an important step to accurately detect particular transportation modes.

The popularization of Location Based Services (LBS) has created new challenges for interaction designers in validating the design of their applications. Existing tools designed to play back GPS location traces data streams have shown potential for testing LBS applications and for supporting rapid and reflective prototyping. However, selecting a useful set of location traces from among a large collection remains a difficult task. In this paper, we present TraceViz, the first system that is aimed specifically at supporting LBS designers in exploring, filtering, and selecting location traces. TraceViz employs dynamic queries and "brushing" to allow LBS designers to flexibly adjust their trajectory filter criteria to find location traces of interest. An evaluation performed with eight LBS designers and developers indicates that TraceViz is helpful for rapidly locating useful traces and also highlights areas for future improvement.