School of Interactive Computing Technical Report Series

Series

School of Interactive Computing Technical Report Series

Permanent Link

https://hdl.handle.net/1853/71008

Series Type

Publication Series

Associated Organization(s)

Organizational Unit

College of Computing

Organizational Unit

School of Interactive Computing

Full item page

Publication Search Results

Now showing 1 - 10 of 28

Planar similarity-motion interpolating three keyframes: Comparative assessment of prior and novel solutions

(Georgia Institute of Technology, 2021) Rossignac, Jarek ; Vinacua, Àlvar

We compare 8 solutions for defining the planar motion of an oriented edge that interpolates 3 keyframes. One contribution is the discovery of several novel solutions, one of which produces what we call a locally-perseverant motion, for which the acceleration of a moving point remains constant in the local (moving) frame. The other contribution is to demonstrate that: (a) many interesting solutions exist, (b) the mathematical and perceived differences between the animations they produce are significant, and (c) these differences may matter for designers and applications. To allow motions that rotate by more than 2π, we represent the 3 keyframes and the moving edge by arrows, each storing the starting-point p of the edge, its length m, and its winding (arbitrary angle) w. Hence, an arrow defines an integer winding-count k (with |w − 2kπ| ≤ π) and a similarity transformation that combines dilation by m, rotation by w − 2kπ, and translation from the origin to p. Our chosen PITA (Planar Interpolation of Three Arrows) solutions are formulated using compositions of linear, polar, or log-spiral interpolations, or using ODEs or logarithms of matrices. We compare these solutions in terms of 11 mathematical properties and also in terms of subjective attributes that may be important for designers. We illustrate differences between our 8 chosen PITAs in 6 use-cases: Keyframe-animation, Variable-width stroke design, Banner deformation, Pattern animation, Motion prediction, and Curve design.
A virtual coach for question asking and enabling learning by reflection in startup engineering

(Georgia Institute of Technology, 2020-12) Goel, Ashok K. ; Hong, Sung Jae ; Kuthalam, Mukundan ; Arcalgud, Arup ; Gulati, Siddharth ; Howe, James ; Karnati, Nikhita ; Mardis, Aaron ; Ro, Jae ; McGreggor, Keith

The Socratic method of teaching engages learners in extended conversations and encourages learning through answering questions, making arguments, and reflecting on the evolving conversation. This method can be a powerful instrument of learning by reflection, especially in domains in which the right answers to open questions are not known in advance such as entrepreneurship. In this paper, we describe an initial experiment in developing AI technology for simulating the Socratic method of teaching in learning about entrepreneurship. When a would-be entrepreneurs creates a business model on the Business Model Canvas (BMC), the AI agent named Errol uses semantic and lexical analysis of the entries on the BMC to ask questions of the students. By attempting to categorize and correct the errors that novices typically make, Errol seeks to accelerate the process by which a novice can start creating more expert-like business models
CvLabs: A Container Based Interactive Virtual Lab for IT Education

(Georgia Institute of Technology, 2020-04) Chen, Geng

With the ongoing trend of digital transformation, the need for more well-trained IT professionals is keep raising, as well as the need for quality IT education. Experiential learning is a key component for learners to practice problem solving and develop skills in many IT courses. Virtual lab is an effective solution to meet the increasing demand on experiential learning environment. However, delivering and managing satisfying virtual labs are technically and economically challenging for instructors, lab designers and educational institutions. This paper presents the design and implementation for CvLabs, a container based interactive virtual lab system. The system can provide learners with ready-to-use virtual learning environments. It also enables learners to share lab sessions to improve collaboration, which is missing in many implementations of virtual lab. CvLabs is built with industrial standard open source container technologies. Container provides a foundation for the delivery of consistent virtual lab environments and improves learning experience. It also simplifies the creation of reusable lab contents, which can significantly reduce the workload of lab instructors and lab designers. The introduction of container ensures CvLabs can be deployed on various infrastructures and scaled to support a large number of learners. By leveraging cloud infrastructure, CvLabs can lower the cost of delivering and managing virtual labs for educators and educational institutions. CvLabs was deployed for a test run with a small group of test users. The system received positive feedback on the learning experience improvement. The preliminary results from this work shows containers is a promising technology for experiential learning in IT education.
Tools for Measuring and Understanding the Proximity of Users to Their Smartphones

(Georgia Institute of Technology, 2020-03) Park, Jung Wook ; Evans, Hayley I. ; Watson, Hue L. ; Abowd, Gregory D. ; Arriaga, Rosa I.

Two studies in ubiquitous computing examined the proximity of users to their smartphones in 2006 and in 2011. Both studies have used a passive data collection tool and the day reconstruction method. Additionally, Dey at al. adopted an online survey to validate their findings with a larger population sample. In 2019, we attempted to revisit this research topic due to the high adoption rate of smartphone and smart- watch. In our replication study, we developed a new passive data collection tool and a novel survey technique, proximity-based ecological momentary assessments. We also adopted the day reconstruction method and online survey utilized in the previous studies. This technical report presents the details of the research tools and techniques used in our study. This technical report is a supplementary material to the published article, "Growing Apart: How SmartDevices Impact the Proximity of Users to Their Smartphones", in IEEE Pervasive Computing.
JobLex: A Lexico-Semantic Knowledgebase of Occupational Information Descriptors

(Georgia Institute of Technology, 2019-07) Saha, Koustuv ; Reddy, Manikanta D. ; De Choudhury, Munmun

Technological advancements in several work sectors have influenced evolution of the landscape of work at an unprecedented speed, leading to the demand of continuous skill development [1,8]. In turn, this interests a number of stakeholders spanning across academia and industry in a number of disciplines including labor economics, who leverage large-scale data available from a variety of offline and online sources (e.g., resumes, job portals, professional social networking such as LinkedIn, search engine, job databases, etc.) [9,11,12]. On these data streams, describing job aspects and skills vary extensively, confounded by factors such as self-presentation, subjective perspectives on soft and hard skills, audience, and intrinsic traits such as personality and mindset [2,4,7,15,17]. Such data analyses require a taxonomy of keywords that are associated with skills per job description or type. However, most databases are only limited — they do not capture variants, typos, abbreviations, or internet slangs that are used on social media or in informal settings [6]. To facilitate research in this space, our work builds on a well-validated dictionary of occupational descriptors (O*Net) to propose a method, and correspondingly a knowledgebase, JobLex of occupational descriptors that can be used in computational social science and organizational studies [13]. We publish both our script and an example lexicon (for Twitter) for purposes of research and practical application. Item Description: This work publishes a methodology to build knowledgebase of occupational information descriptors. We also provide access to the codebase and example lexicon.
Developing a Document Trained Automated Advisor

(Georgia Institute of Technology, 2018-08) Gregori, Eric

This paper covers the development of a system to automatically answer questions about the content of a document. For example, a class syllabus or project specification. The system trains on the document’s content to build a model and answer questions using text from the document.
Jill Watson: A Virtual Teaching Assistant for Online Education

(Georgia Institute of Technology, 2016) Goel, Ashok K. ; Polepeddi, Lalith

MOOCs are rapidly proliferating. However, for many MOOCs, the effectiveness of learning is questionable and student retention is low. One recommendation for improving the learning and the retention is to enhance the interaction between the teacher and the students. However, the number of teachers required to provide learning assistance to all students enrolled in all MOOCs is prohibitively high. One strategy for improving interactivity in MOOCs is to use virtual teaching assistants to augment and amplify interaction with human teachers. We describe the use of a virtual teaching assistant called Jill Watson (JW) for the Georgia Tech OMSCS 7637 class on Knowledge-Based Artificial Intelligence. JW has been operating on the online discussion forums of different offerings of the KBAI class since Spring 2016. By now some 750 students have interacted with different versions of JW. In the latest, Spring 2017 offering of the KBAI class, JW autonomously responded to student introductions, posted weekly announcements, and answered routine, frequently asked questions. In this article, we describe the motivations, background, and evolution of the virtual question-answering teaching assistant.
An Experiment in Teaching Cognitive Systems Online

(Georgia Institute of Technology, 2015) Goel, Ashok K. ; Joyner, David A.

In Fall 2014 we offered an online course CS 7637 Knowledge-Based Artificial Intelligence: Cognitive Systems (KBAI) to about 200 students as part of the Georgia Tech Online MS in CS program. We incorporated lessons from learning science into the design of the project-based online KBAI course. We embedded ~150 microexercises and ~100 AI nanotutors into the online videos. As a quasi-experiment, we ran a typical inperson class with 75 students in parallel, with the same course syllabus, structure, assignments, projects and examinations. Based on the feedback of the students in the online KBAI class, and comparison of their performance with the students in the inperson class, the online course appears to have been a success. In this paper, we describe the design, development and delivery of the online KBAI class. We also discuss the evaluation of the course.
Deliberate Barriers to User Participation on MetaFilter

(Georgia Institute of Technology, 2014) Pileggi, Hannah ; Morrison, Briana ; Bruckman, Amy S.

This descriptive study explores deliberate barriers to user participation on the long-lived discussion site Metafilter.com. Metafilter has been in continuous operation since its founding in 1999, and at the time of this writing has around 12,000 active users. While many newer online sites appear eager to eliminate barriers to participation and recruit as many new members as possible, Metafilter charges a $5 fee to join and has a mandatory one-week waiting period before new users are allowed to post. In this paper, we explore both why these barriers were imposed and why some users choose to surmount the barriers to become members. Our data sources include historical documents posted on the site, interviews with eleven site members, an informal user survey, and an interview with the Matt Haughey, the site’s founder and owner. Implications of these design features are discussed.
EasyZoom: Zoom-in-Context Views for Exploring Large Collections of Images

(Georgia Institute of Technology, 2013) Chen, Jiajian ; Xu, Yan ; Turk, Greg ; Stasko, John T.

Image browsing and searching are some of the most common tasks in daily computer use. Zooming techniques are important for image searching and browsing in a large collection of thumbnail images in a single screen. In this paper we investigate the design and usability of different zoom-in-context views for image browsing and searching. We present two new zoom-in-context views, sliding and expanding views, that can help users explore a large collection of images more efficiently and enjoyably. In the sliding view the zoomed image moves its neighbors away vertically and horizontally. In the expanding view, the nearby images are pushed away in all directions, and this method uses a Voronoi diagram to compute the positions of the neighbors. We also present the results of a user study that compared the usability of the two zoom-in-context views and an overlapping, non-context zoom in the tasks of searching to match an image or a text description, and the task of brochure making. Although the task completion times were not significantly different, users expressed a preference for the zoom-in-context methods over the standard non-context zoom for text-matching image search and for image browsing tasks.