Research projects are listed in alphabetical order by faculty member’s last name. As new proposals arrive, this list will be updated. If you have a great idea for a project that you don’t see listed, please meet with the faculty member most closely interested in that area and propose your idea.
Dr. Michael Bloodgood:
Dr. Bloodgood’s research topics are in machine learning and natural language processing. Possible projects include:
- Stopping Machine Learning. We are researching principled methods for detecting when to stop the collection of additional training data for building machine learning systems.
- Neural Machine Translation. We are researching how to efficiently build state of the art machine translation systems using neural networks and GPUs (graphics processing units).
- Active Learning to Improve Cost-Effectiveness of Annotation. We are researching how to selectively sample data for annotation with the aim of maximizing the benefits received from annotation effort.
- Computer-Assisted Translation Technologies. We are researching how to improve CAT (computer-assisted translation) technologies so that professional translation can be performed faster and with higher quality. For related publications and news articles covering my inventions in this area, please see http://www.tcnj.edu/~bloodgm1/.
Active Learning with Neural Networks. We are researching how to efficiently train neural networks with smaller amounts of training data while not sacrificing any performance.
For related publications and more information about research projects, please see http://www.tcnj.edu/~bloodgm1/.
Dr. Uddipan Das:
Dr. Das conducts research in the area of smart cities focusing on the design and development of data-driven algorithms and systems in smart grids and smart living. In particular, his current research interest includes data management in communication networks of smart grids that can improve reliability and enable better management and control of electric meters in residential units, as well as the design and development of auxiliary wayfinding systems for people with visual impairments for smart living.
Students are encouraged to contact Dr. Das (email: email@example.com) to discuss details of his research projects.
Dr. Deborah Knox:
- Mobile computing for iOS, Android, or non-native programming – No previous mobile application development experience is required. Join a successful mobile app development team!
Possible projects include:
- TCNJ Campus Tour app needs a redesign. Project may integrate of points-of-interest technology.
- Indoor navigation using technology such as beacons or WiFi (indoor positioning systems) to augment navigation in the TCNJ Library
- Music repertoire personal assistant. Explore optical character recognition (OCR) technology and integrate data management and personalization strategies to create a new approach to tracking works performed or under study
- Explore platform independent application development tools (producing non-native apps for wider distribution on any platform)
- Recent products from the TCNJ LakesideApps Development Team, led by Dr. Knox include:
- TCNJ Connect (Version 2 released October 6, 2014)
- TCNJ Library (Version 2 expected Fall 2014)
- TCNJ Campus Tour
- TCNJ Connect (expected Fall 2014)
- TCNJ BookNav (release date pending)
Campus stakeholders test all apps branded with the TCNJ name and campus administrative approval was received prior to public release.
- Big Data and Sensor Based Monitoring – Sensors collect volumes of data, some of which is sensitive. Explore methods of data collection from sensors, handle storage, and perform extraction. Conduct research in current technologies for data masking, which maintains data’s utility, as well as redaction approaches for unstructured data for maintaining privacy. Other projects may be proposed. Other research areas may be explored. Please request an appointment to discuss your interests.
Dr. Jikai Li:
- Secure cloud file storage – This project will try to develop a framework, including encryption process and key management on client side, to encrypt/decrypt the files to/from cloud. We expect the framework will work on a broad range of electronic devices. The software is simple enough to be able reinvented if it is lost. The key management is solid and convenient to users.
- Spoofing caller detection – It is becoming common practice that a hacker/telemarketer to spoof caller ID. This ongoing project will explore the possibility of using tools like Asterisk to detect the spoofing caller ID. Additional info used for detection can be online registration info etc.
Dr. Dimitris Papamichail:
My research focuses on applied algorithms. Projects I undertake usually involve the design and/or implementation of algorithms, occasionally supplemented by web-based interfaces. I’m looking for algorithmically inclined students interested to work on the following topics:
- Synthetic biology – Design algorithms and software to construct novel synthetic genes and protein libraries. Interested students ideally should have taken a course in algorithm design. It would be useful to have some introductory knowledge of biology, but not a requirement.
- Automated music transcription – I am interested in algorithmic techniques for automated transcriptions of music scores from one set of instruments to another. Interested students should have some background in algorithms, programming experience, basic music theory, and playing knowledge of at least one music instrument.
- Other algorithmic projects – I often explore other smaller scale projects involving algorithms and their implementations, drawn primarily from computational biology but also other fields. Feel free to come and ask about available projects, or bring your own ideas involving algorithmic optimization.
Dr. Monisha Pulimood:
My research centers around the confluence of human and artificial intelligence, and I am looking at computational problems that cannot be completely solved by computers alone, but need some intervention by humans. It builds on work I have been doing (with students and other faculty) on how concepts from human computation and collective intelligence can help build more vibrant communities in collaborative environments and make them less time-consuming to maintain. I am particularly interested in investigating leveraging the collective intelligence of large numbers of humans to improve the reliability and integrity of data, and the sustainability of such applications, and to analyze how ideas propagate.
Students will primarily work on CABPortal to continue research, design and development of a web-based portal to make resources associated with the Collaborating Across Boundaries (CAB) model (see my Research page for details) publicly available. We are leveraging concepts from artificial intelligence and social computational systems, such as human computation, crowdsourcing, and the collective intelligence of large numbers of humans, to motivate potential adopters to become active participants in improving the reliability and integrity of the hosted data and resources. CABPortal is designed to be a self-sustaining, collaborative environment that can also track how ideas and content are propagated and the CAB model is adopted. The intent is to eventually make the portal available for faculty at TCNJ and on other campuses to list their own collaborative courses. The genesis of the project was funded through the NSF TUES grant (Award #1141170), and funding continues through the NSF IUSE grant (Award # 1914869), both of which are described on my Research (http://pulimood.pages.tcnj.edu/research/) page.
Dr. Andrea Salgian:
- American Sign Language recognition
- Virtual personal trainer
- Conducting tutor – helping music students learn how to conduct
Dr. Yoon’s research interests span around artificial intelligence, machine learning, computer vision, and multimedia. In general, his research goal is to fill in the gap between individual human decision processing and the group-level behavior, which typically based on distributed or decentralized individual decisions, and eventually devise a generalized computational model to reproduce them. More details on his research can be found on his website. Students who are interested in the problems listed therein are encouraged to contact him for specific project topics they can work on. Motivated students may propose their own machine learning project. Please contact Dr. Yoon for details.
Dr. Sharif Mohammad Shahnewaz Ferdous
Dr. Ferdous’ research interests include Virtual Reality, Augmented Reality, Human Computer Interaction, Game Development, and Interactive Computer Graphics. Research projects include:
- Visual Sickness: The visual sickness project focuses on developing an animated representation of the existing simulator sickness question. The simulator sickness questionnaire is a standard way of measuring a person’s sickness level in an immersive virtual environment. We are developing a visual sickness questionnaire that uses animated emoticons to show the questionnaire’s different symptoms. This will make it easy for children to understand the simulator sickness questionnaire and answer them correctly.
- Wind Immersion: The wind immersion project focuses on generating realistic wind based on user speed in a virtual environment. The objective of this research is to compare the difference in immersion and cybersickness in three conditions – no wind, constant wind and variable wind with respect to navigational speed. Our hypothesis is – wind will have a significant effect on improving immersion and reducing cybersickness in virtual reality. We have designed a prototype using DC fans.
- PepAR: The PepAR project aims to make an augmented reality system that allows users to scan their papers and work on them collaboratively. Users from different places will use a traditional pen to draw on a piece of paper, and our AR system will superimpose other users drawing on top of the paper. This way, collaborative work on an existing hard copy will be much easier for remote users.
- Adaptive HMD Cooling: This project uses the temperature and humidity data inside a virtual reality head-mounted display (HMD) to cool down the inside of the HMD. Our hypothesis is that maintaining a comfortable environment inside the HMD will give users a better experience of virtual reality and make them less sick.
- AR as a learning media for children with Autism Spectrum Disorder: The objective of this research is to compare video and augmented reality and conclude which media is best for learning for children with ASD. Our hypothesis is – Children with ASD learn better in augmented reality than video. We develop augmented reality simulation that teaches the molecule shape using Merge cube. Users can rotate Merge cubes to see different molecules from each angle. We also developed a video version of the application. They have done a pilot study to validate the system.
- Effect of a customizable virtual classroom on learning: The objective of this research is to compare the learning outcome in virtual reality-based the ability to customize different classroom features. Our hypothesis is – customized virtual classroom increases the learning outcome and engagement of students. We developed a virtual reality classroom that has a fully functional virtual teacher, virtual classmates, and other features that users can customize.