Organizational Unit:

School of Interactive Computing

Permanent Link

https://hdl.handle.net/1853/70783

Parent Organization

Organizational Unit

College of Computing

ArchiveSpace Name Record

https://finding-aids.library.gatech.edu/agents/corporate_entities/1113

Full item page

Publication Search Results

Now showing 1 - 10 of 10

BrainBraille: Towards Passive Training in Brain-Computer Interfaces using fNIRS

(Georgia Institute of Technology, 2023-01-18) Gemicioglu, Tan

Amyotrophic Lateral Sclerosis (ALS) is a debilitating movement disability that causes patients to gradually lose their ability to voluntarily control their muscles. In some cases, patients who are "locked-in" are unable to move any muscles, leaving them with no means of communicating with caregivers. Brain-computer interfaces (BCIs) attempt to create a means of communication directly through brain activity, removing the need for movement. BrainBraille is a novel interaction method for BCIs, enabling complex text-based communication using attempted movements with a six-region pseudo-binary encoding. In this dissertation, I explore a wearable BCI using functional near-infrared scanning (fNIRS) to make BrainBraille mobile. In an early study, I show that transitional gestures based on executed movements of two hands can be classified in two participants with up to 93% accuracy. I explore how transitional gestures can benefit BrainBraille by expanding the vocabulary and enabling faster responses. Finally, I evaluate future paths for integrating passive haptic training into BrainBraille to reduce the physical exertion needed to learn a BCI for ALS patients.
Interactive 3D User Interface for Sensor Placement on On-Body locations

(Georgia Institute of Technology, 2022-05) Bhardwaj, Sukriti

IMUTube generates on-body sensor-based human activity recognition data by leveraging the abundance of videos on platforms like YouTube as a dataset for training human activity recognition models. This thesis deals with the development of a user interface for IMUTube which will allow a user, such as a computer vision researcher or healthcare workers, to generate human pose detection data automatically using IMUTube by inputting custom data while hiding the nuances of the state-of-the-art computer-vision and signal-processing based back-end system. Thus, the user interface creates an interactive web application that makes IMUTube accessible to users from different backgrounds and requires no coding experience.
Machine Learning based Procedural Content Generation in Semantic Choreography

(Georgia Institute of Technology, 2020-05) Xiao, Kyle Phillip

BeatMania is a rhythm-action game where players press buttons in response to keysound events to generate music. Rhythm-action game charts (the sequence of keysound events) have traditionally been human authored, since each song level must be creatively organized and correspond an overall pattern or theme. A deep neural network approach is proposed for rhythm-action game chart creation, and a method of level evaluation for co-creative AI is defined. That is, given an arbitrary piece of music, human users can generate BeatMania charts as well as give input to an AI collaborator. The problem is divided into two parts: autonomous chart generation and design interaction. For the chart generation process, a combination of features that include grouping information and audio sample labels are incorporated into an artificial neural network. For the design interaction, principal component analysis is utilized for a proposed reinforcement learning model. The co-creative tool is tested against Markov Chain and LSTM baselines via human trials.
Deep Convolutional Player Modelling on Log and Level Data

(Georgia Institute of Technology, 2018-05) Liao, Nicholas

We present a novel approach to player modeling based on a convolutional neural net trained on game event logs. We test our approach and a hybrid extension over two distinct games, a clone of Super Mario Bros. and Gwario, a human computation version of Super Mario Bros: The Lost Levels. We demonstrate high accuracy in predicting a variety of measures of player experience across these two games. Further we present evidence that our technique derives quality design knowledge and demonstrate the ability to build a more general model.
Analysis of Deviant Opioid Addiction Treatment Communities on Reddit

(Georgia Institute of Technology, 2018-05) Zampieri, Francisco Alejandro

As the opioid epidemic in the US continues, many addicts turn to clinically unverified, non-mainstream, deviant recovery methods to ameliorate the symptoms of withdrawal. In this study, we analyze discussion on the social media site Reddit surrounding these treatments. We apply transfer learning methods to train a classifier highly sensitive to recovery-related posts. Based on network analysis of Reddit communities (known as “subreddits”), we generate a list of subreddits where discussion of deviant addiction treatment methods is taking place. Using word embeddings and the testimony of a practicing opioid addiction clinician, we identify potential alternative opioid addiction treatment methods. Applying the classifier to subreddit post data, we generate a dataset consisting of recovery-related discourse. When applied to these posts, topic modeling methods, such as Latent Dirichlet Allocation (LDA), reveal topics discussed within the context of recovery, such as the lifestyle changes associated with kratom use.
Examining the Potential of Low-Cost Virtual Reality for Low-Resource Learning Contexts

(Georgia Institute of Technology, 2018-05) Vishwanath, Aditya

We present a qualitative evaluation of a field deployment of the Google Expeditions Virtual Reality (VR) Toolkit at an after-school learning center for children from marginalized backgrounds in Mumbai, India. The lowering costs of VR technology make it especially attractive to study the feasibility of this and analogous toolkits in low-resource environments in and outside of India. Using a combination of interviews and observations, we demonstrate the feasibility of integrating VR into the classroom setting we study. We also illustrate the technical, social, and information dimensions that are critical for supporting this integration and present recommendations for deploying VR systems in other, diverse low-resource learning environments, seeking to motivate future examination of this space.
Designing an Effective Interactive E-book for Computer Science Education

(Georgia Institute of Technology, 2015-01-28) Moore, Steven James

There’s a need to develop more computer science teachers around the world. As massive open online courses have been failing, an interactive e-book used for distance learning might fulfill this need. This research seeks to determine what makes for good usability in a computer science e-book and measuring if teachers learn using them. The study investigates participants’ preferences for interactive computer science e-book designs and usability, in an attempt to review and develop guidelines for educational e-book creation. Comparisons and preferences were made between three interactive educational computer science e-books and specific multimedia widgets within them. Based on the reported findings, some general guidelines were suggested for increasing usability within interactive e-books and ways to enhance their educational value. The results of the study suggest that interactive educational e-books are an effective alternative to current distance learning methods for teaching computer science.
Designing a character avatar model for the Mermaids MMO

(Georgia Institute of Technology, 2010-12-20) Ead, Samer Omar

This paper describes the technique formed for the creation of an efficient, simply rigged, customizable mermaid avatar model for the Mermaids massively multiplayer online game (MMOG). Our goal was to improve the in game customization of the player s 3D mermaid model, while maintaining rendering efficiency. We devised a procedure that utilizes the iterative nature of design without sacrificing the scientific and technical aspects of the project. Our procedure begins by a method known as Partitioning where we break down the model s body into distinct sub-models. During development, this partitioning allowed us to focus on smaller concise areas of interest, whereas during game-play this change granted the player greater strands of customization freedom. Since the model relied on a skeleton for its animations, it s partitioning required Skeletal Reformations to reassess the control scheme of the rig over the sub-models. In this method, individual sub-skeletons were designed to provide increased local control over their respective sub-models in contrast to the global control that the previous rig allowed. The sub-skeletons were then joined together forming a combined and complete skeletal rig for the mermaid model. We iterated through the previous methods refining their procedures in efforts of Balancing Customizability with Efficiency , which in turn provided us with the results of our novel technique. Our technique utilizes innovative methods that localize skeletal control over respective sub-models in a novel way, which allows increased customizability with limited costs to efficiency.
The Online Community Grid Volunteer Grid Computing with the Web Browser

(Georgia Institute of Technology, 2010-05-12) Miller, Daniel Menachem

Current community grid projects, such as IBM's World Community Grid (www.worldcommunitygrid.org), have successfully developed standalone applications to connect thousands of clients to one huge network of users. Each client donates their machine's idle time to compute mathematical operations to help solve multiple humanitarian projects requiring massive amounts of data to be computed (examining tissue microarrays, human proteome folding, ect). These projects have been relatively successful; however, there are multiple design problems that hinder a multitude of users to join the network. These problems include yet are not limited to: ● Required user registration and email verification ● Once a user registers, a large standalone application to transmit and receive data from a central server is required to be downloaded and installed ● The program runs as a background process and is only active when a user's computer is idle ● The program may not work due to firewall restrictions on the client machine This research, as shown by prototype, eliminates all of these barriers. No registration, no installation, no required idleness, and no firewall issues. According to Adobe/Macromedia, roughly 96-98% of all home computers have the Flash Player installed. This figure is greater than any operating system, browser, program, or other virtual machine (including Java). The purpose of this research project is to determine whether it is possible to create a grid community on the Internet utilizing browser technologies such as Flash and or AJAX technology.
Community Mosaic: Finding Ways to Eat More Healthfully in a Low-Income African American Community

(Georgia Institute of Technology, 2010-05-12) Sharma, Mansi

Community Mosaic is a system to promote healthy eating amongst low-income African American people. This thesis analyzes a study I conducted that encourages people to talk about what they think are the barriers against as well as the resources for trying to eat healthfully. The results of this study will be useful for the low income African American community as well as the HCI (Human Computer Interaction) community. More specifically, the results will facilitate towards finding ways to eat more healthily due to the correct understanding of the barriers against and available resources for eating healthfully in a community. The African American community is known to have more diet-related health problems than other ethnic communities in the US; by understanding the reasons behind these problems the goal is to help minimize them.