Research Projects

Below is a list of all projects available for summer 2021. Please review these prior to applying; in the application you will be asked to select your top three projects of interest. 
 

Accelerating Innovation Through Analogy Mining

Lead Mentor: Hyeonsu Kang

Project Description: We are looking for students for building prototype interactive systems to accelerate the rate of scientific innovations through mining analogies. Students will work with a graduate student mentor and a faculty advisor at HCII to gain hands-on experience in building interactive visualizations and developing and applying cutting edge natural language processing models. As a member of this project, you will play a key role in defining, designing, developing, and evaluating interactive visualizations and algorithms involving the challenging dataset of scientific text. You will contribute to advancing techniques for finding deep structural (analogical) relations between scientific concepts that go beyond simple keyword matching based on surface similarity. You will also contribute to scaling these techniques to millions of scientific papers to support interactive visualizations. To evaluate the visualizations, we will conduct user studies measuring how such interfaces might support scientists’ creativity.

Skills or Requirements:

  • Interest and experience in building Web-based interactive systems.
  • Prior knowledge in Natural Language Processing technologies is preferred but not required.
  • We use D3.js for data visualization, and Tensorflow and Apache Beam for developing and deploying machine learning models.

Research Areas in Which This Project Falls:

  • Applied Machine Learning
  • Artificial Intelligence (AI)
  • Data Visualization
  • Methods
  • Scientific Collaboration
  • User Experience (UX)
Accessibility and Autonomous Vehicles

Lead Mentor: Patrick Carrington

Project Description: Autonomous vehicles have the potential to be designed to increase the mobility of those who have physical, sensory, or cognitive disabilities. We are working to develop new design ideas for how to make autonomous cars best serve those with disabilities. To develop new solutions, we are working with members of the disability community and transportation advocates to understand people's mobility needs, current challenges, and hopes for the future. We are looking for student researchers interested in conducting user research, design, and prototyping around making autonomous cars accessible.

Skills or Requirements:

  • Experience conducting user studies
  • Qualitative research (interviews, surveys, focus groups)
  • Participatory design
  • Some experience with electronics
  • Basic experience with programming
  • Some Fabrication experience

Research Areas in Which This Project Falls:

  • Accessibility
  • Artificial Intelligence (AI)
  • Design Research
  • Inter of Things (IoT)
  • Sensors
  • Service Design
  • User Experience (UX)
ADAMITE: Using Annotations to Facilitate Learning Across Developers

Lead Mentor: Brad Myers

Project Description: When learning to use a new piece of software, developers typically spend a large amount of time reading, understanding and trying to answer questions with online materials. To help developers keep track of important, confusing, and useful information with the intent of sharing their learning with future developers, we developed a social annotation tool, Adamite, a Google Chrome extension. For more information, you can read our recently-submitted paper: https://horvathaa.github.io/public/publications/adamite-submission-deano.... While Adamite is successful as an extension, we want you to help us refine Adamite’s current features, and design and build new features depending upon your own interests. Possible project focuses may include: extending Adamite to work in a code editor like Visual Studio Code or on a mobile device, adding new features to make annotations even easier and more beneficial to the author and for subsequent users such as using intelligent techniques to cluster related annotations, studying developers’ actual usage of annotations during a software learning task, or porting Adamite for entirely new domains beyond programming, like shopping. Working on this project may result in being an author on a publication at a prestigious conference such as CHI, UIST, or CSCW, and we are hoping to release Adamite as an open source project for general use.

Skills or Requirements:

  • Required Skills: Some web development experience (e.g., one completed introductory web development course) OR some design and prototyping experience (e.g., experience using Figma or Adobe Creative suite tools) OR experience performing user studies (e.g., interviews, survey design, A/B lab studies)
  • Performing qualitative data analysis (e.g., grounded theory)
  • Preferred skills: Experience using React

Research Areas in Which This Project Falls:

  • Crowd Sourcing
  • End-User Programming
  • Tools
Advancing Fairness in AI with Human-Algorithm Collaborations

Lead Mentor: Haiyi Zhu

Project Description: Artificial intelligence (AI) systems are increasingly used to assist humans in making high-stakes decisions, such as online information curation, resume screening, mortgage lending, police surveillance, public resource allocation, and pretrial detention. While the hope is that the use of algorithms will improve societal outcomes and economic efficiency, concerns have been raised that algorithmic systems might inherit human biases from historical data, perpetuate discrimination against already vulnerable populations, and generally fail to embody a given community's important values. Recent work on algorithmic fairness has characterized the manner in which unfairness can arise at different steps along the development pipeline, produced dozens of quantitative notions of fairness, and provided methods for enforcing these notions. However, there is a significant gap between the over-simplified algorithmic objectives and the complications of real-world decision-making contexts. This project aims to close the gap by explicitly accounting for the context-specific fairness principles of actual stakeholders, their acceptable fairness-utility trade-offs, and the cognitive strengths and limitations of human decision-makers throughout the development and deployment of the algorithmic system.

To meet these goals, this project enables close human-algorithm collaborations that combine innovative machine learning methods with approaches from human-computer interaction (HCI) for eliciting feedback and preferences from human experts and stakeholders. There are three main research activities that naturally correspond to three stages of a human-in-the-loop AI system. First, the project will develop novel fairness elicitation mechanisms that will allow stakeholders to effectively express their perceptions on fairness. To go beyond the traditional approach of statistical group fairness, the investigators will formulate new fairness measures for individual fairness based on elicited feedback. Secondly, the project will develop algorithms and mechanisms to manage the trade-offs between the new fairness measures developed in the first step, and multiple existing fairness and accuracy measures. Finally, the project will develop algorithms to detect and mitigate human operators' biases, and methods that rely on human feedback to correct and de-bias existing models during the deployment of the AI system.

Skills or Requirements:

  • Preferred but not required: Working with the project team to plan and conduct research studies, including interviews, user studies, surveys, design workshops, or behavioral experiments.
  • Preferred but not required: Analyzing and interpreting data collected from research studies, in collaboration with other project team members.
  • Preferred but not required: Ideating and designing new tools to improve fairness in machine learning practice.

Research Areas in Which This Project Falls:

  • Applied Machine Learning
  • Artificial Intelligence (AI)
  • Social Computing
  • Societal Problems
Advancing Metamaterials by exploring novel structures, developing design tools and fabrication methods

Lead Mentor: Alexandra Ion

Project Description: We are looking to push the boundaries of mechanical metamaterials by unifying material and device. Metamaterials are advanced materials that can be designed to exhibit unusual properties and complex behavior. Their function is defined by their cell structure, i.e., their geometry. Such materials can incorporate entire mechanisms, computation, or re-configurable properties within their compliant cell structure, and have applications in product design, shape-changing interfaces, prosthetics, aerospace and many more.

In this project, we will develop design tools that allow novice users and makers to design their own complex materials and fabricate them using 3D printing or laser cutting. This may involve playfully exploring new cell designs, creating novel application examples by physical prototyping and developing open source software.

Skills or Requirements:

  • CS skills: software development, background in geometry, optimization, and/or simulation
  • 3D modeling basics (CAD tools, e.g., Autodesk Fusion 360 or similar)
  • Basic knowledge of classical mechanics or material science

You don’t have to cover all skills, since this will likely be a group project. We are looking for diverse teams with complementary skills.

Research Areas in Which This Project Falls:

  • Computer Aided Manufacturing
  • Enabling Technologies
  • Tools
AI and User Experience Design

Lead Mentor: John Zimmmerman

Project Description: People love products and services that use artificial intelligence (AI) to make the product work better. Today, more and more companies are searching for ways to do this. From spam filters that save people time and attention to recommenders that make it easier to find something of interest to conversational agents that offer a more natural way to interact with a computer to fully functioning smart homes and driverless cars, AI can make things better.

Unfortunately, UX designers, the people most often asked to come up with new ideas for products and services, really struggle when trying to innovate with AI. These professionals often fail to notice the many simple ways that AI can make people’s interaction better. In addition, when they do try to envision new things, they most often generate ideas for things that cannot be built. Our work focuses on helping UX designers to become better at envisioning what AI can do and then communicating their ideas to development teams.

Our work addresses this challenge in three ways. First, we are making resources to help designers better understand AI’s capabilities and its dependency on labeled datasets. Second, we are making new design tools that scaffold designers in thinking through what interaction with a probabilistic system might be like, a system that can make inference errors. Third, we are working with professional designers working in industry to better understand their work practices and to identify the best time and place for them to use the resources and the tools.

We are looking for research assistants to help us with our research. This work will involve:
1. Designing user interfaces that make our AI resources available to designers. These include a taxonomy of AI capabilities and a collection of AI interaction design patterns.
2. Designing user interfaces for new tools that help designers recognize when they should search for opportunities to use AI to enhance their designs.
3. Conducting interviews and participating in workshops with professional designers who want to get better at working with AI.

Students working on this project will learn about AI capabilities from a UX design perspective, and they will develop resources for designers to leverage AI opportunities in their work.

Skills or Requirements:

  • Design Research (e.g., user interviews, design workshops, affinity diagrams)
  • User interface design (sketching and prototyping of nobel UIs)
  • Interest in Human-AI Interaction
  • Strong organizational skills, reliable, self-motivated
  • Education in data science, analytics, data mining, and or experience working with user telemetry data

Research Areas in Which This Project Falls:

  • Artificial Intelligence (AI)
  • Design Research
  • Service Design
  • Tools
  • User Experience (UX)
AI Driven Tutoring System Authoring Tools

Lead Mentor: Daniel Weitekamp

Project Description: Looking for a student interested in interaction design and/or AI to help with the Apprentice Learner Framework, and Smart Sheet. The goal of these two project are to build an AI system that can be used for rapid Intelligent Tutoring System authoring. The Apprentice Learner learns to solve problems via a humans' demonstrations and correctness feedback and in turn produces a set of rules that can be used as a tutoring system for students. SmartSheet aims to use AL in conjunction with handwriting recognition to make handwriting recognition capable tutoring systems built via a tablet and stylus based interface. We are also interested in hosting students interested in expanding the Apprentice Learner Framework in general, including for the purposes of cognitive modeling.

Skills or Requirements:

  • Requirement: Solid programming skills (mostly Python, Javascript, would also be helpful)
  • Preferred: Some interest/experience with machine learning
  • Optional: UX design/implementation skills

Research Areas in Which This Project Falls:

  • Applied Machine Learning
  • Artificial Intelligence (AI)
  • Intelligent Tutoring Systems
  • User Experience (UX)
Augmented Reality Meets Computational Interaction 

Lead Mentor: David Lindlbauer

Project Description: Augmented Reality and Virtual Reality offer interesting platforms to re-define how users interact with the digital world. It is unclear, however, what the requirements in terms of usability and interaction are to avoid overloading users with unnecessary information. In this project, we will build on existing machine learning approaches such as saliency prediction that leverage insights into human visual perception. The goal is to create computational approaches to improve the applicability and usefulness of AR and VR systems.

Skills or Requirements:

  • Strong technical background 
  • Some experience with 3D editors (e.g. Unity, Unreal) and 3D programming
  • Some familiarity with Computer Vision 

Research Areas in Which This Project Falls:

  • Applied Machine Learning
  • Augmented Reality (AR)
  • Context-Aware Computing
  • Virtual Reality (VR)
Data Visualization Toolkits 

Lead Mentor: Dominik Moritz

Project Description: Vega-Lite is a high-level grammar of interactive graphics. It provides a concise, declarative JSON syntax to create an expressive range of visualizations for data analysis and presentation. It is used by thousands of data enthusiasts, ML scientists, and companies around the world. We have a number of projects around adding new features to the visualization toolkits that are going to be part of the open-source tool. Please take a look at https://docs.google.com/document/d/1fscSxSJtfkd1m027r1ONCc7O8RdZp1oGABwca2pgV_E and the issue trackers for some specific project ideas we could work on.

Skills or Requirements:

  • Experience with web-development (JavaScript) and Git
  • Experience with data visualization, TypeScript, D3, and Vega are a plus

Research Areas in Which This Project Falls:

  • Data Visualization
  • Tools
Decimal Learning Game Designer

Lead Mentor: Bruce McLaren

Project Description: In this project you will help in the design of visual and experimental aspects of a decimal number learning game, Decimal Point (http://www.cs.cmu.edu/~bmclaren/projects/DecimalPoint/), to prepare it for new classroom studies. You will work with a professor and researchers who do learning science studies in middle school classrooms. You should have design skills, both to help with new artwork in the game but also to prepare for the classroom studies. An ideal candidate will be someone with a psychology and/or design/art background.

Skills or Requirements:

  • Design skills
  • UI skills
  • Game design skills

Research Areas in Which This Project Falls:

  • Education
  • Games
  • Learning Sciences and Technologies
  • Social Good
  • User Experience (UX)
Decimal Learning Game Programmer

Lead Mentor: Bruce McLaren

Project Description: In this project you will write and revise code to alter an existing decimal number learning game, Decimal Point (http://www.cs.cmu.edu/~bmclaren/projects/DecimalPoint/), to prepare it for new classroom studies. You will work with a professor and researchers who do learning science studies in middle school classrooms. You’ll learn about those studies, as well as practice important professional technical skills, such as using source code repositories and engaging in good software engineering practice. Preferred, but not necessary, is that you will have skills in HTML5/CSS3/JavaScript. Familiarity with Angular or AngularJS is also desirable.

Skills or Requirements:

  • Programming skills (e.g., Python, Java)
  • Knowledge of HTML5/CSS3/JavaScript
  • Knowledge of Angular or AngularJS
  • A desire to work with a fun research team!

Research Areas in Which This Project Falls:

  • Education
  • Games
  • Intelligent Tutoring Systems
  • Learning Sciences and Technologies
  • Social Good
  • User Experience (UX)
Enabling Adaptive Practice of Prior Knowledge towards Promoting Education Equity

Lead Mentor: Vincent Aleven

Project Description: Students will learn the best when offered the timely, proper scaffoldings that correspond to their current knowledge level. With AI-based intelligent tutoring systems that can calculate students current skill mastery of certain knowledge, it is possible to offer personalized, adaptive practice based on each students’ prior knowledge. This may better prepare students up to the next-to-be-learned knowledge, reduce the time of them struggling unproductively or practicing knowledge they have already mastered, and increase their learning efficiency through more personalized, focused and targeted practice. Such adaptive practice of prior knowledge also holds the potential to bridge the knowledge gap between different students, and promote education equity. In this project, we are interested in exploring the design space of building such a system/ algorithm.

Skills or Requirements:

  • Preferred skills in web-based technologies development
  • Preferred skills HCI/design skills, especially related to web-applications
  • Preferred skills in AI/machine learning/data mining

Research Areas in Which This Project Falls:

  • Artificial Intelligence (AI)
  • Design Research
  • Education
  • Intelligent Tutoring Systems
  • Leaning Sciences and Technologies
  • Social Good
  • User Experience (UX)
Engaging educational technology for middle-school students

Lead Mentor: Vincent Aleven

Project Description: In research studies, we found out that Lynnette, an AI-based tutoring system for middle-school equation solving, is very effective in helping students learn. To take Lynnette to the next level, we are now working to make it more engaging, by gamification elements such as a space theme, a badge system, achievements, and a narrative context. We also added a drag-and-drop interaction format for equation solving, for variety and smooth interactions. We would like to make Lynnette even more engaging. We have a variety of ideas, including personalized dashboards, culturally-adaptive story problems, and adapting instruction to social and meta-cognitive factors (e.g., sense of belonging in classroom, self-efficacy). We are open to other ideas as well. The work involves design brainstorming, co-design with middle-school students, prototyping, and trying out prototypes with students.

Skills or Requirements:

  • Preferred skills: Design/HCI, prototyping, web development
  • Experience with game design, educational technology, and designing for teenagers is not required but would be considered a plus

Research Areas in Which This Project Falls:

  • Design Research
  • Education
  • Games
  • Intelligent Tutoring Systems
  • Learning Sciences and Technologies
  • User Experience (UX)
Fabricating Morphing Materials

Lead Mentor: Lining Yao

Project Description: Leveraging stimuli responsive actuators and smart materials to develop wearables and second-skin applications that sense and actuate. We look to investigate novel fabrication processes, building novel manufacturing tools, and looking into new structures and mechanisms of physical prototypes. Students are encouraged to look for a balance of impactful application and fundamental research.

Skills or Requirements:

  • We are looking for candidates who have hands-on design and making experiences.
  • If you have digital fabrication experiences, or hands-on craft/making projects, please send an email and share your portfolio to Prof. Lining Yao: liningy [at] andrew.cmu.edu

Research Areas in Which This Project Falls:

  • Computer Aided Manufacturing
  • Enabling Technologies
  • Wearables
Fuse: a tool for making sense of complex information

Lead Mentor: Niki Kittur

Project Description: Over a trillion hours per year are spent searching for and making sense of complex information. For a lot of this "sensemaking," search is just the beginning: it’s also about building up your landscape of options and criteria and keeping track of all you’ve considered and learned as you go. We've found through more than a decade of research that existing tools don't support this messy process. So we are building a tool that does.

Skills or Requirements:

  • Looking for undergraduates with background and interest in one or more of:
    • Design, both visual and interaction
    • UX research
    • Customer discovery and lean methods
    • Front end programming (especially with React/Firebase experience)

Research Areas in Which This Project Falls:

  • Design Research
  • Service Design
  • Tools
  • User Experience (UX)
Human-AI co-orchestration of dynamically-differentiated, collaborative classrooms

Lead Mentor: Vincent Aleven

Project Description: Our research experience in K-12 classrooms equipped with AI-based tutoring systems has shown that, even though the software is coaching each student individually with adaptive guidance, teachers still play an enormous role in student learning. For example, we have seen teachers in these classrooms team up students on the fly (e.g., a student who has learned a lot already with one who is struggling), for brief one-on-one extra aid. For teachers to be most effective, however, they must be able, in real time, to see how their students are doing (struggling, disengaged, very far into covering the learning objectives,, etc.). This information will enable them to take action to aid those who need help. We are creating tools by which middle-school teachers can effectively orchestrate activity in these classrooms without taking their attention away from the students. The tools use wearable and mobile devices with artificial intelligence to turn the voluminous data generated by the software into actionable diagnostic information and recommendations for teachers. In a new project, we plan to provide tutoring software that supports both individual and collaborative problem-solving by students, to enhance the effectiveness of the spontaneous teaming up of students we have observed. We seek student interns interested in any of the many efforts needed to make these technologies work, such as testing and refining collaborative ITSs with students; doing data mining to create new learning analytics that inform teachers about how their students are doing; and design-based research for designing and prototyping intuitive, comprehensible visualizations for teachers, for different hardware options, and trying them out with teachers.

Skills or Requirements:

  • HCI/design skills (preferred)
  • AI/machine learning/data mining (preferred)
  • Web technologies (preferred)

Research Areas in Which This Project Falls:

  • Applied Machine Learning
  • Artificial Intelligence (AI)
  • Data Visualization
  • Design Research
  • Intelligent Tutoring Systems
  • Learning Sciences and Technologies
  • User Experience (UX)
Interactive Motivational Interviewing Training

Lead Mentor: Laura Dabbish

Project Description: In this project we are developing an interactive prototype for motivational interviewing training using a human-centered design approach. Motivational Interviewing (MI) is an effective therapeutic technique to support behavior change, but training is often time consuming and its effectiveness diminishes over time.

Students on this project will work in a team to design and develop a prototype for time effective interactive MI training useful both for initial training and refreshers. Our prototype will build on our research to identify the unique challenges nurses face in learning MI and that MI newcomers struggle with building rapport, analyzing the problem, and promoting readiness for change during therapeutic interactions.

Skills or Requirements:

  • Interest in interactive prototype development, interaction design, psychology and/or mental health
  • Web programming (front-end or back-end)
  • Familiarity with inVision, Sketch, or other graphic design tools
  • Familiarity with Python or Javascript

Research Areas in Which This Project Falls:

  • Design Research
  • Education
  • Games
  • Healthcare
  • Social Computing
Learning to See, Seeing to Learn

Lead Mentor: Marti Louw

Project Description: Macroinverbrates.org: The Atlas of Common Freshwater Macroinvertebrates of the Eastern United States has been successfully launched as a definitive teaching and learning collection and online guide for freshwater macroinvertebrate identification with annotated key diagnostic characters marked down to family and genus for the 150 most commonly used taxa in citizen science water quality assessment and education/ This year we are completing the design, development, usability testing, evaluation and release of a fully downloadable mobile app for both Android and iOS operating systems that includes a Aquatic Insect Field Guide, Interactive Identification key (Orders) mode, and self-practice quizzing and games feature.

Skills or Requirements:

  • Design and evaluation skills (interviewing, usability testing, qualitative analysis skills)
  • Copy writing and tutorial video production
  • Technical Skills: android and iOS development in React Native.

Research Areas in Which This Project Falls:

  • Design Research
  • Education
  • Learning Sciences and Technologies
  • Scientific Collaboration
Mixed-Reality Learning

Lead Mentor: Nesra Yannier

Project Description: NoRILLA is a project based on research in Human Computer Interaction at Carnegie Mellon University. We are developing a new mixed-reality educational system bridging physical and virtual worlds to improve children's STEM learning and enjoyment in a collaborative way. It uses depth camera sensing and computer vision to detect physical objects and provide personalized immediate feedback to children as they experiment and make discoveries in their physical environment. NoRILLA has been used at many school districts, Children's Museum of Pittsburgh, Carnegie Science Center and informal play spaces like IKEA and Bright Horizons. Research with hundreds of children has shown that it improves children's learning by 5 times compared to equivalent tablet or computer games. We have recently received an NSF (Advancing Informal STEM Learning) grant in collaboration with Science and Children's Museums around the country to expand our Intelligent Science Stations/Exhibits and develop the AI technology further. Responsibilities will include taking the project further by developing new modules/games, computer vision algorithms and AI enhancements on the platform and deployment of upcoming installations, as well as participating in research activities around it.

Skills or Requirements:

  • The project has both software and hardware components.
  • Familiarity with computer vision, Processing/Java, game/interface development and/or robotics is a plus.

Research Areas in Which This Project Falls:

  • Artificial Intelligence (AI)
  • Augmented Reality (AR)
  • Education
  • Games
  • Learning Sciences and Technologies
  • Societal Good
Non-programmer authoring of AI-based tutoring software (CTAT)

Lead Mentor: Vincent Aleven

Project Description: Past research shows that AI-based tutoring systems--software that coaches students step-by-step as they try to solve complex problems--can be very helpful aids to learning, for example in middle school and high school mathematics learning. The effectiveness of tutoring software has been shown in many different subject areas, and some have reached commercial markets and are in daily use in many schools. Inspired by these results, we have developed authoring tools that make the creation of tutoring software much easier--in many cases, it can be done entirely without programming, opening the door to a wide range of authors. Instructors in fields such as math and physics have used our tools to create tutoring software for their classes. We are now creating a new version of our authoring tools with the knowledge we have gained over the 20 years since their first release. We seek to make them intuitive, clear, efficient and readily available--without requiring complicated installation or advance instruction. We want to improve both the creation of the student-facing user interface and the specification of the tutoring knowledge and behavior behind it. We want to support teams of teachers collaboratively creating these self-coaching, self-grading online activities that they might assign instead of conventional homework.

Skills or Requirements:

  • The work involves user-centered design of authoring tools, web programming, building tool prototypes, and trying them out with tutor authors of various background.

Research Areas in Which This Project Falls:

  • Artificial Intelligence (AI)
  • Education
  • End-User Programming
  • Intelligent Tutoring Systems
  • Learning Sciences and Technologies
  • User Experience (UX)
Opportunistic health sensing using wearables

Lead Mentor: Mayank Goel

Project Description: A key trend emerging from the popularity of smart, mobile devices is the quantified-self movement. The movement has manifested into the prevalence of two kinds of personal wellness devices: (1) fitness devices (e.g., FitBit), and (2) portable and connected medical devices (e.g., Bluetooth-enabled blood pressure cuffs). The fitness devices are seamless, very portable, but offer low-fidelity information to the user. They do not generate any medically-relevant data. We are currently working on building personal medical devices that are as seamless to use as a FitBit but generate medically-relevant data. We are looking for students to contribute to various aspects of this project. Depending on their interest, the students can help in building and prototyping the mobile app, hardware device, or they can contribute to the signal processing and machine learning component.

Skills or Requirements:

Some subset of these skills would be useful:

  • Programming experience in Java and/or Python
  • Mobile programming
  • Machine learning
  • Hardware prototyping

Research Areas in Which This Project Falls:

  • Applied Machine Learning
  • Healthcare
  • Internet of Things (IoT)
  • Sensors
  • Societal Problems
  • Wearables
Peekaboo: Providing Architectural Support for Building Privacy-sensitive Smart Home Apps

Lead Mentor: Haojian Jin

Project Description: While privacy is an important element for smart home products, it takes a great deal of effort to design and develop features to support end-user privacy. These individually built features result in distributed and non-consistent interfaces, further imposing challenges for user privacy management. Peekaboo is a new IoT app development framework that aims to make it easier for developers to build privacy-sensitive smart home apps through the intermediary of a smart home hub, while simultaneously offering architectural support for building centralized and consistent privacy features across all the apps. At the heart of Peekaboo is an architecture that allows developers to factor out common data preprocessing functions from the cloud service side onto a user-controlled hub and supports these functions through a fixed set of reusable, chainable, open-source operators. These operators then become the common structure of all the Peekaboo apps. Privacy features built on these operators become native features in the Peekaboo ecosystem.

Skills or Requirements:

  • Node-JS
  • Python
  • Docker

Research Areas in Which This Project Falls:

  • End-User Programming
  • Internet of Things (IoT)
  • Security and Privacy
  • Sensors
Personalized Learning²

Lead Mentor: Ken Koedinger

Project Description: Personalized Learning² (PL²) is an initiative addressing the opportunity gap for marginalized students through personal mentoring and tutoring with artificial intelligence-powered learning software. (personalizedlearning2.org)

Skills or Requirements:

  • The student researcher will contribute to designs that enhance the general usability of the PL² web application
  • The student researcher will need to draw on their UI/UX and Communication Design skills and prior experience to create mock-ups to be presented for team approval
  • Ideally the candidate will also be able to conduct user interviews to determine their needs and contribute to the creation of demo videos
  • Front-end programming experience (HTML, CSS, Javascript) is a nice-to-have but not required

Research Areas in Which This Project Falls:

  • Design Research
  • Learning Sciences and Technologies
  • User Experience (UX)
Smart Spaces for Making: Networked Physical Tools to Support Process Documentation and Learning

Lead Mentor: Marti Louw

Project Description: Multiple positions are open for summer undergraduate research assistants on an NSF-funded project Smart Spaces for Making: Networked Physical Tools to Support Process Documentation and Learning. Candidates are sought for two roles:1.Design Research(skills: conducting interviews, qualitative research analysis)2.Technical Prototype Development(skills: hardware prototyping, server-side programming) Research assistants will work with an interdisciplinary team of design researchers, technology developers and learning scientists from CMU’s HCII, School of Architecture and the University of Pittsburgh’s Learning Research and Development Center (LRDC) to develop smart documentation tools to support learning practices in creative, maker-based studio environments. Students will participate in design research and development activities working with project site partners including Quaker Valley High School, CMU’s IDeATe program, and AlphaLab Gear’s Startable Youth program. For more details on the project: https://smartmakingtools.weebly.com

Skills or Requirements:

  • Design Research (skills: contextual inquiry, interviewing, design probes and qualitative research analysis)
  • Technical Prototype Development(skills: hardware prototyping, server-side programming)

Research Areas in Which This Project Falls:

  • Education
  • Internet of Things (IoT)
  • Learning Sciences and Technologies
  • Methods
Social Cybersecurity Games and Interventions

Lead Mentor: Laura Dabbish

Project Description: This project leverages theory from psychology and known social influence principles to improve cybersecurity behavior and enhance security tool adoption. Students on this project will work on designing and developing web-based security-related mini games or interventions that incorporate cybersecurity training and information into people’s everyday workflows. They may also have an opportunity to conduct evaluations of these mini games or interventions.

Skills or Requirements:

  • Interest in interaction design, psychology and/or cybersecurity
  • Web programming (front-end or back-end)
  • Familiarity with inVision, Sketch, or other graphic design tools
  • Familiarity with Python or Javascript

Research Areas in Which This Project Falls:

  • Design Research
  • Games
  • Security and Privacy
  • Social Computing
  • Societal Problems
Social Programmable Robots - CS Ed

Lead Mentor: Amy Ogan

Project Description: This project focuses on creating a culturally-responsive programming curriculum for underrepresented middle schoolers of color in a coding camp. REU students will participate in creating, refining, and assessing a programming curriculum. We look forward to working with students who would like to pursue this area of work. It is favorable if REU students have: an excitement to promote diversity in and accessibility to computing, a strong interest in and/or prior experience in data analysis, motivation and time management skills, and some experience in programming.

Skills or Requirements:

  • Data Analysis
  • Python Programming
  • Learning Science

Research Areas in Which This Project Falls:

  • Education
  • Human-Robot Interaction (HRI)
  • Learning Sciences and Technologies
  • Social Good
Social Programmable Robots Design

Lead Mentor: Amy Ogan

Project Description: In this project, we explore how social robots can be used in out-of-school learning environments to empower Black, Latinx, and Native American middle school girls in computer science. We are using a culturally-responsive computing paradigm to examine how to reflect the learner's identity in the robot, towards the goal of improving learning outcomes, and self-efficacy in computer science. We are looking for motivated REU students, interested in exploring methods of endowing the robot with social characteristics and building rapport between the robot and the learner. REU students will work on data collection and/or analysis, and interaction design. Qualified students should have experience in programming/computer science as well as design/HCI or human-robot interaction.

Skills or Requirements:

  • Design
  • Python Programming
  • User Research

Research Areas in Which This Project Falls:

  • Design Research
  • Education
  • Human-Robot Interaction (HRI)
  • Learning Sciences and Technologies
  • Social Good
Studying Social Influences on Security and Privacy Behavior

Lead Mentor: Laura Dabbish

Project Description: Research on the human factors of cybersecurity often treats people as isolated individuals rather than as social actors within a web of relationships and social influences. We are developing a better understanding of social influences on security and privacy behaviors across contexts. Students on this project will conduct interviews and surveys on people's cybersecurity behaviors in different types of relationships. They may also have the opportunity to develop and prototype new system design ideas based on research findings.

Skills or Requirements:

  • Coursework in psychology, design or cybersecurity
  • Interest in developing interviewing and survey design skills
  • Experience with statistical analysis techniques or packages such as R
  • Interest in interaction design

Research Areas in Which This Project Falls:

  • Design Research
  • Security and Privacy
  • Social Computing
  • User Experience (UX)
Supporting Learners across School and Home

Lead Mentor: Amy Ogan

Project Description: This project (SEME) hopes to design a feasible and sustainable chatbot to mentor teachers in order to support them in their implementation of teacher training methods in rural Ivory Coast. We will be building the chatbot on Facebook and deploying it at scale for the Fall of 2021. (Details: https://seme-cmu.github.io/seme-web/). We are looking for REU students interested to work on the design, data analysis, and development aspects of the project.

Skills or Requirements:

  • Design
  • Data analysis
  • Development

Research Areas in Which This Project Falls:

  • Developing World
  • Learning Sciences and Technologies
  • Social Good
  • Societal Problems
Supporting Responsible use of Algorithmic Decision-support in High-stakes Contexts

Lead Mentor: Ken Holstein

Project Description: Today, algorithmic decision-support systems (ADS) guide human decisions across a growing range of high-stakes settings – from predictive risk models used to inform child welfare screening decisions, to AI-based classroom tools used to guide instructional decisions, to data-driven decision aids used to guide mental health treatment decisions. While ADS hold great potential to foster more effective and equitable decision-making, in practice these systems often fail to improve, and may even degrade decision quality. Human decision-makers are often either too skeptical of useful algorithmic recommendations (resulting in under-use of decision support) or too reliant upon erroneous or harmfully biased recommendations (adhering to algorithmic recommendations, while discounting other relevant knowledge). To date, scientific and design knowledge remains scarce regarding how to foster appropriate levels of trust and productive forms of human discretion around algorithmic recommendations.

In this project, we will investigate how to support more responsible and effective use of ADS in real-world contexts where these systems are already impacting the lives of children and families (e.g., mental healthcare, child welfare, and K-12 education). We will create new interfaces and training materials to help both practitioners and affected populations in these contexts decide: (1) when and how much to adhere to algorithmic recommendations, and (2) how to act upon or communicate (dis)trust. In addition, we will conduct experiments to evaluate the impacts different interface and training designs have on human–algorithm decision-making.

Skills or Requirements:

  • Experience conducting research with human subjects (e.g., user studies, design workshops, field studies, experiments) is preferred, but not required.
  • Experience with front-end design and/or development is a preferred, but not required.
  • Interests and/or background in any of the following areas are a plus: design, cognitive science, anthropology, decision science, statistics, artificial intelligence, machine learning, psychology, learning sciences.

Research Areas in Which This Project Falls:

  • Applied Machine Learning
  • Artificial Intelligence (AI)
  • Data Visualization
  • Design Research
  • Education
  • Ethics
  • Healthcare
  • Learning Sciences and Technologies
  • Social Good
  • Societal Problems
The Future of Work in the Hospitality Industry

Lead Mentor: Jodi Forlizzi

Project Description: This project is seeking an undergraduate researcher with skills in HCI, design research, and/or social computing to conduct research on the impact of automation in the hospitality industry. This role is primarily dedicated to study design, data collection, and project management related to research on the Future of Work. This collaboration creates a unique opportunity for innovation focused on identifying challenges hospitality workers experience due to emerging technology, defining the pipeline of innovations in hospitality, evaluating workforce issues including training challenges and deficiencies, and examining policy questions and options.

Skills or Requirements:

  • Design
  • Social Computing
  • Miro
  • Organize and run online meetings

Research Areas in Which This Project Falls:

  • Design Research
  • Social Computing
  • Social Good
  • Societal Problems
Towards Design Practices Accessible for People with Disabilities

Lead Mentor: Jeffrey Bigham

Project Description: People with disabilities regularly encounter accessibility barriers in their daily lives. For example, you might notice crosswalks that do not have a curb cut, so a wheelchair user would have difficulty crossing the street, or you might watch videos posted online that do not have captions accompanying spoken dialogue so a Deaf or hard of hearing viewer cannot read along. As such, the field of accessibility aims to relieve such barriers through technology development. Unfortunately, like in the general public, a lot of tools and techniques that accessibility professionals use to develop accessible solutions are not actually accessible for people with disabilities to use. For example, tables in a makerspace may not be adjustable so a wheelchair user can work comfortably, and many physical computing tutorials assume that a user can use vision to read schematics and other instructions, leaving these activities difficult for blind and low vision people to learn. But in parallel, people with disabilities regularly adapt their worlds in response to access barriers, coming up with all sorts of creative solutions, about which prospective REU students can learn more by reading the podcast and its resources linked below. Given mismatches between the ways people with disabilities express their creativity and the tools, processes, and cultures that underpin accessibility research, many people with disabilities are not recognized for their creativity and encounter extreme difficulties working in design and technical fields. As such, this project will continue ongoing research to learn from current creative practices undertaken by people with disabilities. From these insights, REU students will conduct research to identify how the field of accessibility can support these creative and highly personalized adaptations through accessible tools, instructions, processes, cultures, and more. Activities will be tailored to the progress of the project at the REU’s outset and the student’s interests and talents. They may include but are not limited to, qualitative surveys, interviews or focus groups, prototyping tools and techniques, or developing documentation of accessible practices. https://www.mapping-access.com/podcast/2020/6/8/contra-podcast-episode-2...

Skills or Requirements:

  • Qualitative methods such as interviewing, surveys, focus groups
  • Low and high fidelity prototyping
  • Fabrication and physical computing
  • Experiences doing activist or community-based research or other projects

Research Areas in Which This Project Falls:

  • Accessibility
  • Design Research
  • Social Good
  • Societal Problems