Users on Twitter, a microblogging service, started the phenomenon of adding tags to their messages sometime around February 2008. These are distinct from those in other Web 2.0 systems because users less likely index for later retrieval. We compare tagging patterns Twitter with Delicious show that behavior is different its conversational, rather than organizational nature. use mixed method statistical analysis and an interpretive approach study phenomenon. find more about filtering directing...

Users frequently modify a previous search query in hope of retrieving better results. These modifications are called reformulations or refinements. Existing research has studied how web engines can propose reformulations, but given less attention to people perform reformulations. In this paper, we aim understand searchers refine queries and form theoretical foundation for reformulation. We study users' reformulation strategies the context AOL logs. create taxonomy refinement build high...

Understanding how people interact with search engines is important in improving quality. Web typically analyze queries and clicked results, but these actions provide limited signals regarding interaction. Laboratory studies often use richer methods such as gaze tracking, this impractical at scale. In paper, we examine mouse cursor behavior on engine results pages (SERPs), including not only clicks also movements hovers over different page regions. We: (i) report an eye-tracking study showing...

Video games are primarily designed for the players. However, video game spectating is also a popular activity, boosted by rise of online sites and major gaming tournaments. In this paper, we focus on spectator, who emerging as an important stakeholder in games. Our study focuses Starcraft, real-time strategy with millions spectators high level tournament play. We have collected over hundred stories Starcraft spectator from sources, aiming diverse group possible. make three contributions...

Past studies of user behavior in Web search have correlated eye-gaze and mouse cursor positions, other lines research found interactions to be useful determining intent relevant parts pages. However, are not all the same; different types patterns exist, such as reading, hesitating, scrolling clicking, each which has a meaning. We conduct study with 36 subjects 32 tasks determine when gaze aligned, thus position is good proxy for position. effect time, patterns, user, task on gaze-cursor...

We introduce SearchGazer, a web-based eye tracker for remote web search studies using common webcams already present in laptops and some desktop computers. SearchGazer is pure JavaScript library that infers the gaze behavior of searchers real time. The tracking model self-calibrates by watching interact with pages trains mapping features to locations page elements on screen. Contrary typical information retrieval, this approach does not require purchase any additional specialized equipment,...

We present SleepCoacher, an integrated system implementing a framework for effective self-experiments. SleepCoacher automates the cycle of single-case experiments by collecting raw mobile sensor data and generating personalized, data-driven sleep recommendations based on collection template created with input from clinicians. The guides users through iterative short to test effect their sleep. evaluate in two studies, measuring frequency awakenings, self-reported restfulness, onset latency,...

Search trails mined from browser or toolbar logs comprise queries and the post-query pages that users visit. Implicit endorsements many can be useful for search result ranking, where presence of a page on trail increases its query relevance. Follow-ing requires user effort, yet little is known about benefit obtain this activity versus, say, sticking with clicked jumping directly to destination at end trail. In paper, we present log-based study estimating value following. We compare...

Web search components such as ranking and query suggestions analyze the user data provided in click logs. While this is easy to collect provides information about behavior, it omits interactions with engine that do not hit server; these logs omit users' cursor movements. Just clicks provide signals for relevance results, hovering scrolling can be additional implicit signals. In work, we demonstrate a technique extend models of user's result examination state infer document relevance. We...

Understanding the impact of individual and task differences on search result page examination strategies is important in developing improved engines. Characterizing these effects using query click data alone common but insufficient since they provide an incomplete picture behavior. Cursor- or gaze-tracking studies reveal richer interaction patterns are often done small-scale laboratory settings. In this paper we leverage large-scale rich behavioral log a naturalistic setting. We examine...

We provide the first solid evidence that Chinese superstitious beliefs can have significant effects on house prices in a North American market with large immigrant population. Using real estate data close to 117,000 sales, we find houses address number ending “4” are sold at 2.2% discount and those “8” 2.5% premium comparison other addresses. These price found either neighborhoods higher than average percentage of residents, consistent cultural preferences, or repeated transactions,...

In our age of ubiquitous digital displays, adults often read in short, opportunistic interludes. this context Interlude Reading , we consider if manipulating font choice can improve adult readers’ reading outcomes. Our studies normalize size by human perception and use hundreds crowdsourced participants to provide a foundation for understanding, which fonts people prefer make them more effective readers. Participants’ speeds (measured words-per-minute (WPM)) increased 35% when comparing...

Abstract Many studies have examined how the 2019 Coronavirus Disease (COVID-19) has impacted sleep health. Early evidence suggests that lockdown policies worldwide led to changes in timing, duration, and quality; however, few attempted look at longer-term effects across multiple countries a large data set. This study uses self-reported from 64,858 users of Sleep As Android smartphone application around world over 24-month period 2020. We found significant but modest increase time bed (TIB),...

Abstract The study of expertise is difficult to do in a laboratory environment due the challenge finding people at different skill levels and lack time for participants acquire mastery. In this paper, we report on two studies that analyze naturalistic gameplay data using cohort analysis better understand how relates practice habit. Two cohorts are analyzed, each from games ( Halo Reach StarCraft 2 ). Our work follows progression through 7 months matches holistic perspective, but also...

This paper introduces two novel algorithms for learning behaviors from human-provided rewards. The primary novelty of these is that instead treating the feedback as a numeric reward signal, they interpret form discrete communication depends on both behavior trainer trying to teach and teaching strategy used by trainer. For example, some human trainers use lack indicate whether actions are correct or incorrect, interpreting this accurately can significantly improve speed. Results user studies...

When visualizing data with uncertainty, a common approach is to treat uncertainty as an additional dimension and encode it using visual variable. The effectiveness of this depends on how the variables chosen for representing other attributes interact influence user's perception each We report user study graph edge when associated main attribute are visualized simultaneously two separate variables. covers four that commonly used line graphical primitives: lightness, grain, fuzziness,...

Parallel browsing describes a behavior where users visit Web pages in multiple concurrent threads. browsers explicitly support this by providing tabs. Although parallel is more prevalent than linear online, little known about how perform activity. We study the use of through log-based millions and present findings on their behavior. identify power law distribution browser metrics comprising "outclicks" tab switches, which signify degree browsing. find that switch tabs at least 57.4% time,...

Existing sleep-tracking apps and devices provide simple descriptive statistics or generic recommendations for everyone. In this work, we aim to leverage cohort-based sleep data improve an individual's sleep. We report a 4-week study (N = 39) conducted compare three alternatives: 1) no recommendation, 2) general 3) using six quality metrics. For the were generated based on "similar users" about 40 million events from Microsoft Band users. Our results indicate that systems health can prompt...

Smartphone augmented reality (AR) lets users interact with physical and virtual spaces simultaneously. With 3D hand tracking, smartphones become apparatus to grab move objects directly. Based on design considerations for interaction, mobility, object appearance physics, we implemented a prototype portable tracking using smartphone, Leap Motion controller, computation unit. Following an experience prototyping procedure, 12 researchers used the help explore usability issues define space. We...

The message a designer wants to convey plays pivotal role in directing the design of an infographic, yet most authoring workflows start with creating visualizations or graphics first without gauging whether they fit message. To address this gap, we propose Epigraphics, web-based system that treats "epigraph" as first-class object, and uses it guide infographic asset creation, editing, syncing. text-based recommend visualizations, graphics, data filters, color palettes, animations. It further...

Search engines return ranked lists of Web pages in response to queries. These are starting points for post-query navigation, but may be insufficient search tasks involving multiple steps. trails mined from toolbar logs start with a query and contain visited by one user during navigation. Implicit endorsements many can enhance result ranking. Rather than using solely improve ranking, it also worth providing trail information directly users. In this paper, we quantify the benefit that users...

Self-experiments allow people to investigate their own individual outcomes from behavior change, often with the aid of personal tracking devices. The challenge is design scientifically valid self-experiments that can reach conclusive results. In this paper, we aim understand how novices run when they are provided a structured lesson in experimental design. We conducted study on self-experimentation two cohorts students, where total 34 students performed self-experiment choice. first cohort,...