Chair: Amanda Williams, Concordia University, Canada
Representing Our Information Structures for Research and for Everyday Use
Contribution & Benefit: To realize a scientific inquiry of personal information management (PIM), researchers need methods for representing and measuring information structure. These methods, with small extension, have direct application to end users.
Abstract » We argue for a methodology and supporting infrastructure that promotes a cross-study investigation of information structure to advance the science of personal information management. Moreover, we observe that the infrastructure to support a methodology of scientific inquiry may have direct application to users as they struggle to manage their information. Research on information structure reaches towards a new age in information management wherein organizing information structures grow and change over time based on the internal needs of their owners and not the external demands of tools.
User-Driven Collaborative Intelligence – Social Networks as Crowdsourcing Ecosystems
Contribution & Benefit: Proposes Collaborative Intelligence as a subdiscipline of CHI to evolve platforms for problem-solving by harnessing next generation hybrids of crowd-sourcing and social networks to develop Vernor Vinge’s landmark “singularity” concepts
Abstract » Vernor Vinge proposed, “In network and interface research there is something as profound (and potentially wild) as Artificial Intelligence.” How, in this 2012 Centenary of Alan Turing, can we explore that wild CHI opportunity to create futures of intelligence? User experience data can co-evolve synergies across computer data processing and human capacity for pattern recognition, developing collaborative intelligence applications that engage distributed creativity, processing crowd-sourced analytics to plan and track projects, so that data gathered, bottom-up, can improve decision-making.
Thin Slices of Interaction: Predicting Usersʼ Task Difficulty within 60 sec.
Contribution & Benefit: This study shows that the users’ experienced task difficulty while interacting with a photocopier can be predicted from the automatic video coding of Activity and Emphasis of movement.
Abstract » We report on an exploratory study where the first 60 seconds of the video recording of a user interaction are used to predict the user’s experienced task difficulty. This approach builds on previous work on “thin slices” of human-human behavior, and applies it to human-computer interaction. In the scenario of interacting with a photocopy machine, automated video coding showed that the Activity and Emphasis predicted 46.6% of the variance of task difficulty. This result closely follows reported results on predicting negotiation outcomes from conversational dynamics using similar variables on the speech signal.
Citeology: Visualizing Paper Genealogy
Contribution & Benefit: Presents Citeology, a interactive system to explore the relationships between papers through their use of citations. The full CHI and UIST paper database is used as an example corpus.
Abstract » Citeology is an interactive visualization that looks at the relationships between research publications through their use of citations. The sample corpus uses all 3,502 papers published at ACM CHI and UIST between 1982 and 2010, and the 11,699 citations between them. A connection is drawn between each paper and all papers which it referenced from the collection. For an individual paper, the resulting visualization represents a “family tree” of sorts, showing multiple generations of referenced papers which the target paper built upon, and all descendant generations of future papers.
Mining Whining in Support Forums with Frictionary
Contribution & Benefit: Describes a technique for extracting standardized problem statements from support forums on the web. Mozilla designers and support staff believe it could be useful for prioritizing design decisions.
Abstract » Millions of people request help with software in support forums, creating a massive repository of user experiences ripe for mining. We present Frictionary, a tool for automatically extracting, aggregating, and organizing problem described in support forums, enabling timely problem frequency and prevalence metrics. We applied it to 89,760 Firefox support requests from 4 sources gathered over 10 months. Interviews with the Firefox principal designer and support lead suggest that Frictionary could be a useful tool for prioritizing engineering efforts, but that the extraction would need to be more precise to be useful.