Research projects

Kevin Nam, Mark Ackerman | Publication (Group’07)

Keywords: organizational memory, collective intelligence, knowledge management, online communities, collaborative software, user interfaces, information visualization

A standard problem with online discussions is that once use has ceased either by deadline or by neglect, a site is often a bramble of ideas and topics, too large and unwieldy for its information to be successfully reused. A process of filtering, structuring and organizing of the information, or the process of distillation is needed.     

Two design principles, which we call incremental diagenesis and incremental summarization, help human editors flexibly distill the informal information. Our system, Arkose, is built as a demonstration of these principles, providing the necessary tools for distillation. These tools include a number of visualization mechanisms (Keyword Farm and Author Network) and information retrieval mechanisms, as well as an authoring tool and a navigator for the information space. They support a gradual increase in the order and reusability of the information space and allow various levels of intermediate states of a distillation.

Our current implementation of the Arkose system demonstrates that such a system is feasible although complex.

Some technical details:

Arkose is entirely written in JAVA.  It uses prefuse toolkit for the underlying data visualization, and lucene engine for text analyses and search capabilities.

-

-

-

♦ QuME

Expertise Location and Maintenance System in Online Help Communities

(*this reflects the project description on the SocialWorlds group website.)

-

-

♦ TopicGraph Editor

Graphical Topic Map Authoring Tool

Kevin Nam, Rob Farrell (@ IBM Watson Research)

Keywords: topic maps, authoring, ontology, RDF/OWL, graphical editor framework, rich client platform

Topic maps provide a way to describe the content in file systems or databases, plus relationships that can cross the traditional boundaries between hierarchical organizations of topics.  Topic maps encoded in RDF (Resource

Description Framework) can provide possible paths suitable for searching, browsing, and assembling content.  Dynamic Learning Experience (DLE) is a web-based learning tool that uses topic maps to assemble learning content into individualized courses.  End user authors need a simple and usable tool for creating and editing topic graphs for DLE.

TopicGraph Editor is built in Eclipse using GEF (Graphical Editor Framework) and Rich Client Platform.  It allows a user to interactively assemble topics with relationships, and describes the graph in an RDF format.  Its notable features include flexible addition and deletion of topics and relationships, visual interaction with a map, and easy integration with a reasoning engine.

-

-

♦ Distributed social gaming over distance

Archer Batcheller, Brian Hilligoss, Kevin Nam, Emilee Rader, Marth Rey-Babarro, Xiaomu Zhou| publication: CHI’07

Keywords: social gaming over network, user experiments, new technology, large display

Video connections can establish a media space in which games may be played, just as people play games while collocated. Experiments with participants playing the game ‘Mafia’ indicate that people in a video condition have similar levels of satisfaction, fun, and frustration, to those that play while collocated. This finding holds for both those with prior experience using video systems and those without, suggesting it is not merely a “novelty effect.” Results differ about whether there exist differences in focus of attention, suspicion/trust, and pointing for people playing the game while using a video system. Implications for both fun and work uses of video are suggested.

-

-

♦ iDiag

(*this page reflects the project description on the SocialWorlds group website.)

♦ Pedro

Kevin Nam, Mark Ackerman

Keyword: bio-informatics, distributive resource management, classifier, user interfaces, web 2.0, ontology   

Pedro is a distributed resource discovery/management system over the World Wide Web.  There are many independent, but related resources scattered around on the Web but without an easy way to connect them for a useful purpose.  For example, in the bio-informatics world, researchers used to make a list of links to bio-information tools for their own use.  These pages are in simple HTML, and are static and not maintained over time.  Several links become invalid after a while, but no real way to fix them other than through manual editing.  A beginning researcher who just entered the bio-informatics world has little idea where to locate these tools, what are good tools, and what tools she may need for her research purpose.

Pedro gives a new life to the static and scattered information by connecting them and creating new values.  It consists of:

1) Crawler/ harvester for resource discover: an intelligent web crawler finds similar and related resource pages created independently by researchers, universities, and labs.

2) Classifier: once the crawler/harvester collects resources, a classifier sorts them into related groups.

3) Resource connector: different types of resources are described in RDF (Resource Description Framework), and their use cases are populated.  For example, a bio-informatics scientist who needs to run a specific analysis on certain types of gene sequence may need X, Y tools in Z format.  These use cases are accumulated and rated by peer users in order to create best of practice scenarios.

4) Customized user output and interfaces: allows a combination of different resources that suits an individual user’s needs.

Technical details:  The  Crawler/harvester and user interfaces are written in JAVA, resource ontologies are visually represented using the prefuse visualization toolkit.  Custom output and web interfaces are written in HTML, and Javascript library ExtJS.

-

-

♦ Privacy in ubicomp

Kevin Nam, Atul Prakash, Mark Ackerman

Keywords: ubiquitous computing, privacy policy, user agent, meta-policy, mobile devices, user interfaces

In this project, I created a simple user tool that would semi-automatically set privacy policy settings for a mobile device in a ubiquitous computing environment.

The Computer Science building at the University of Michigan has a sensor network that can track a user’s movement inside the building.  A user can set her privacy settings in order to control who can see her location and at what granularity (e.g., building level, room level, (x,y,z) level).  For example, she can specify her friend may see her location throughout the day, but her advisor may only see her during the business hours.  However, a user’s situation is often dynamic and rapidly shifting, and it would be burdensome for a user to manually set privacy policies for every situation.

We ameliorate the problem by allowing a software agent to know certain context about a user’s situation.  The contextual data may come from a user’s personal application such as a calendar, instant messenger, or email, and a user can set a meta-privacy policy that is easier to create and understand.  For example, when there’s a meeting at 1 pm in the building, participants of the meeting can see each other’s location 5 minutes before and after the meeting time.  By setting the meta-policies, many low-level privacy policies can be semi-automatically set, reducing the user’s burden.

Technical details: User interfaces for meta-policy settings are written in JAVA.  The underlying mechanism that connect the sensor network and user applications is written in Perl.

-

-

♦ Social agents in a forum environment

Kevin Nam, Mark Ackerman

Keywords: social agent system, virtual community, blackboard architecture,
latent semantic indexing

Virtual communities, like any community of people, often have trouble regulating their participants. These problems can range from people being
disruptive, people attacking others verbally, to people abusing group norms.
It may even include people misposting, placing objects in the wrong places,
or otherwise acting erroneously.

This project’s approach includes the design and implementation of an agent-based system written in

Java that helps facilitate and regulate online social spaces appropriately and also helps maintain a pleasurable environment for users. The system comprises of a number of agents that work collaboratively through a blackboard architecture. Each agent looks for a specific problem,  and new agents can be built and added to the system as required. This  system as a whole is not designed to completely eliminate the need for human regulator, but rather to help reduce human intervention in regulating online communities.

The agent system is written in JAVA.  It utilizes the SlashDot web forum as a test bed.  At its core is a blackboard architecture that acts as a central repository for knowledge gathered by many agents, with each agent trying to contribute to a solution to a given problem.  It also implements and utilizes the latent semantic analysis of the posting contents.

-

-

♦ Autominder UI

Kevin Nam, Martha Pollack

Autominder is an “intelligent cognitive orthotic system for people with memory impairment.”  (See this paper for details.)  Using a range of Artificial Intelligent techniques and scheduling algorithms, Autominder models user behaviors and provides adaptive and personalized reminders.

I participated in creating user interfaces in JAVA for the Autominder system, which allowed a user to interact with the scheduler system.  This was a summer research intern position.

-

-

♦ Understanding Reuse of Used Computing Devices (Shaping of Sustainability of PDAs)

Jina Huh, Kevin Nam, Nikhil Sharma | submitted to CHI 2010

Keywords: Sustainable interaction design, sustainability, reuse

We report out findings on the adoption practices of used personal digital assistants (PDAs) to inform reuse of outdated computing products.  Out interviews with 12 eBay users who bought used PDAs showed a variety of ways in which users unknowingly supported sustainability.  This allowed us to re-examine sustainability as something that is dynamically and arbitrarily shaped by the users and not just dependent on the sustainable feature of the product.  We provide design implications for supporting users’ shaping of sustainability.

(more details to come pending the publication.)