10 EECS Undergraduate Students in Haoqi Zhang’s DTR Course Applied for & Received Undergraduate Research Grant Funding

This is the highest number of students supported by a single faculty member in any department across the University.

Photo by Matthew Zhang

In Winter 2016, 10 students in Prof. Haoqi Zhang’s EECS 395, 495: Design, Technology, and Research (DTR) course  applied for Undergraduate Research Grant (URG) funding and all 10 received the funding for 2015-16.

Congrats to Katie George, Greg Kim, Nikhil Pai, Shannon Nachreiner, Ryan Madden, Katherine Lin, Kapil Garg, Jon Rovira, Sarah Lim, and Christina Kim. This is the highest number of students supported by a single faculty member in any department across the University.

Academic Year URGs provide up to $1,000 for your research expenses to do an independent academic or creative project in any field. The project is connected to an independent study or thesis seminar.

Photo by Matthew Zhang

DTR course background

DTR is one of, if not the largest, undergraduate independent research programs on campus and is taught out of the Garage on Fridays from 1-4pm, in the Makerspace room. 20 undergrads and 2-3 grad students receive mentorship and support in independent research projects each quarter, repeated quarterly. Students lead research projects in social and crowd computing, cyberlearning, human computer interaction, and artificial intelligence. 

"DTR's model and approach for doing research through a learning community can fundamentally change how faculty build their research labs while advancing undergraduate and graduate research training in STEM through organizational processes, community structures, and cyberlearning technologies", said Prof Zhang.

DTR Course Facts and Stats:

- Students are almost all CS majors, yet 40% of DTR students are women. Retention beyond the first quarter is near 100%.

- They've hosted 5 quarter-long DTR studios, through which students iteratively designed, built, and tested 15 new socio-technical systems.

- DTR student papers have been accepted for publication at premier conferences in HCI and won 1st, 2nd, and 3rd place at the ACM CHI and ACM Grace Hopper Student Research Competitions.

- DTR students are receiving full-time and internship positions at top tech companies (Google, Apple, Yelp, LinkedIn, Vox Media, Microsoft, …)

- DTR has been funded on a Murphy Society Award for the past 2 years.

DTR students have applied for and received $30,000+ in outside funding for their projects (themselves!). Prof. Zhang has supported 23 DTR students applying for URG funding (or over one third of the overall CS total in the past 6 years). In this past Academic Year URG cycle (2015-16), 10 out of 11 of his students applying for URG funding received funding, an unprecedented success rate and the highest number of students supported by a single faculty member in any department across the University. 

Photo by Matthew Zhang

Project Descriptions

Katie George, Greg Kim, Nikhil Pai [BB] ----- Action Plans

This proposal addresses a critical gap in learning resources and technologies for supporting authentic practice in learning web and software development. While many learning resources and platforms teach syntax and help learners get started, beginning and novice web developers often struggle when transitioning to working on their (first) personal projects. Working on a real project requires understanding how to structure problem solving in ways that aren’t addressed by existing resources and difficult without the help of more experienced peers and mentors. Currently, it is rare to find tutorials addressing learners’ exact projects, and generating a full-fledged tutorial for a single learner is prohibitively expensive.

Nikhil, Greg, and Katie’s proposed work on Action Plans attempts to bridge this gap by providing beginners with customized high-level tutorials (action plans) that more experienced web developers can create. Their ambitious goal is to have “people, with the use of algorithms, […] quickly generate action plans from isolated sources to accomplish complex learning goals.” If successful, their project can fundamentally transform personalized learning in web development and other complex domains in scalable, affordable, and accessible ways.

Shannon Nachreiner, Ryan Madden [BB] ---------- Collective Experiences API

With the Collective Experiences API, Ryan and Shannon are conceptualizing of new ways to connect people beyond social media that make it as easy for people to share and join in an experience together over space and in real-time as it is for us to share information via social media today. The proposed work has the potential to be truly transformative in how we experience others and the world we are in.

A core part of the technology that Shannon and Ryan envision is the User Query engine, which allows an experience creator or organizer to specify with a very simple domain specific language, rich conditions over who are ready and interested in joining in an experience. These conditions identify people in real-time and space and allows for opportunistic interactions whereby potentially large numbers of people sharing some conditions (e.g., all happen to be watching the sunset) can share their experiences in a way that none of them (and many others) couldn’t have experience alone.

Katherine Lin [SC] -------- Habitsourcing

Citizen science plays a crucial role in helping scientists study the natural world, yet challenges in scaling participation in citizen science beyond hobbyists interested in collecting ecological data can significantly limit the number of people contribution and thus the scenarios through which citizen science can achieve good data quality and coverage that permit the desired analysis.

Katherine’s work in habitsourcing seeks to resolve this core challenge by creating interactions that engage order of magnitude more participants in citizen science, by relying not only on hobbyists but engaging large numbers of people practicing their personal routines such as going for a relaxing walk or a run. She is developing interaction techniques and design principles that has the potential to fundamentally change the scale and efficacy of citizen science and the number of people contributing. Katherine recently submitted a paper on her work to ACM UIST 2016.

Kapil Garg [OTG] ------ Low-Effort Sensing for Crowdsourcing

Kapil’s project examines opportunities for using low-effort interactions afforded by the latest mobile and wearable devices for the purpose of communitysensing, that is, the reporting, tracking, and monitoring of public infrastructure by everyday citizens to help governments and policy makers make informed decisions (e.g., think 3-1-1 as an example). Applications such as the widespread monitoring of public infrastructure have the potential to save governments hundreds of millions of dollars in repair and maintenance by helping agencies detect problems early and prioritize fixes.

Kapil’s research questions first examine whether low-effort interactions afforded by Apple Watch and Force Touch on newer iPhones facilitate and encourage low-effort contributions for on-the-go crowds to contribute to sensing efforts. Further, on the motivational side, his research will examine whether access to data can promote further contributions. On the algorithmic side, Kapil is examining methods for indirectly coordinating small low-effort contributions so that collectively, they help to achieve a larger sensing goal without imposing a burden on any contributor.

Jon Rovira [RALE] ----------- Relevant Help from the Professionals: Intelligent Code Matching

Through pilot user surveys and follow-up interviews, Jon identified a critical problem facing many learners who are attempting to learn front-end web development with tutorials that get them up and running quickly but skip foundational computer science knowledge that hinder their growth to becoming better programmers. Jon’s proposed solution is to fill this knowledge gap by exposing such front-end developers to relevant, professional code samples that apply advanced concepts in JavaScript. This approach allows learners to learn by doing — which is important for learner motivation and for getting hands-on experience quickly — while still providing opportunities for learners to reflect on their learning and to examine alternative, professional approaches for implementing similar functionality.

Jon’s work will also extend our ability to practice web foraging; advancing previous work by Brandt et al. which enabled developers to forage tutorials and reference examples, Jon’s work seeks to open up opportunities for eliciting examples from across the open Web.

Sarah Lim and Christina Kim [RALE] --------  Dynamic Tutorials: Creating Readily­ Available Worked Examples for Novice Programmers

Novice programmers increasingly turn to online resources such as tutorials for learning programming concepts and getting help to make progress on their personal projects.

While existing online tutorials demonstrate the “how to do,” most do not provide the “when to do” intuition necessary for effective knowledge mapping to related contexts. Sarah and Christina are advancing an alternative approach, which they call Dynamics Tutorials, that seeks to overcome such learning barriers by (1) web­sourcing for example materials, (2) repeating concepts to promote analogical encoding, and (3) promoting authentic learning through real­ world examples.

Photo by Matthew Zhang