Vivekanandan Suresh Kumar

Prof. Dr. Vivekanandan Suresh Kumar

School of Computing and Information System, Faculty of Science and Technology, Athabasca University, Edmonton, AB, Canada

Research Focus & Metrics

Academic Career Length

30Years (1997-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
15 September 2020 in International Journal of Artificial Intelligence in Education
Verified
Journal Article
Open access
PDF file
VKVivekanandan Suresh Kumar
Vivekanandan Suresh Kumar
DBDavid Boulanger
David Boulanger
This article investigates the feasibility of using automated scoring methods to evaluate the quality of student-written essays. In 2012, Kaggle hosted an Automated Student Assessment Prize contest to find effective solutions to automated testing and grading. This article: a) analyzes the datasets from the contest – which contained hand-graded essays – to measure their suitability for developing competent automated grading tools; b) evaluates the potential for deep learning in automated essay scoring (AES) to produce sophisticated testing and grading algorithms; c) advocates for thorough and transparent performance reports on AES research, which will facilitate fairer comparisons among various AES systems and permit study replication; d) uses both deep neural networks and state-of-the-art NLP tools to predict finer-grained rubric scores, to illustrate how rubric scores are determined from a linguistic perspective, and to uncover important features of an effective rubric scoring model. This study’s findings first highlight the level of agreement that exists between two human raters for each rubric as captured in the investigated essay dataset, that is, 0.60 on average as measured by the quadratic weighted kappa (QWK). Only one related study has been found in the literature which also performed rubric score predictions through models trained on the same dataset. At best, the predictive models had an average agreement level (QWK) of 0.53 with the human raters, below the level of agreement among human raters. In contrast, this research’s findings report an average agreement level per rubric with the two human raters’ resolved scores of 0.72 (QWK), well beyond the agreement level between the two human raters. Further, the AES system proposed in this article predicts holistic essay scores through its predicted rubric scores and produces a QWK of 0.78, a competitive performance according to recent literature where cutting-edge AES tools generate agreement levels between 0.77 and 0.81, results computed as per the same procedure as in this article. This study’s AES system goes one step further toward interpretability and the provision of high-level explanations to justify the predicted holistic and rubric scores. It contends that predicting rubric scores is essential to automated essay scoring, because it reveals the reasoning behind AIED-based AES systems. Will building AIED accountability improve the trustworthiness of the formative feedback generated by AES? Will AIED-empowered AES systems thoroughly mimic, or even outperform, a competent human rater? Will such machine-grading systems be subjected to verification by human raters, thus paving the way for a human-in-the-loop assessment mechanism? Will trust in new generations of AES systems be improved with the addition of models that explain the inner workings of a deep learning black box? This study seeks to expand these horizons of AES to make the technique practical, explainable, and trustable.

10 December 2014 in Journal of Computers in Education
Verified
Journal Article
Open access
PDF file
VKVivekanandan Suresh Kumar
Vivekanandan Suresh Kumar
DBDavid Boulanger
David Boulanger
JSJeremie Seanosky
Jeremie Seanosky
KIKinshuk
Kinshuk
KPKarthikeyan Panneerselvam
Karthikeyan Panneerselvam
TSThamarai Selvi Somasundaram
Thamarai Selvi Somasundaram
A smart learning environment is characterized by the key provision of personalized learning experiences. To approach different degrees of personalization in online learning, this article introduces a framework called Smart Competence Analytics in LEarning (SCALE) that tracks finer-level learning experiences and translates them into opportunities for customized feedback, reflection, and regulation. The SCALE framework is implemented in four layers: the sensing layer, the analysis layer, the competence layer, and the visualization layer. The sensing layer provides the datasets to support context-awareness through state-of-the-art sensing technologies. The analysis layer, by the means of powerful code analysis tools, derives performance metrics (e.g., learner coding metrics) which serve as input to the competence layer to identify proficiency levels of learners. Finally, a learning analytics dashboard called MI-DASH (visualization layer) allows interaction with performance and competence metrics. The current SCALE system has been used in a study to track the habits, performances, and competences of novice programmers. Growth of coding competences of first-year engineering students has been captured in a continuous manner. Students have been provided with customized feedback to optimize their learning paths. The article describes the analytics-based approach pursued in the study and highlights key findings.

1 January 2005
Verified
Conference Paper
Closed access
VKVivekanandan Suresh Kumar
Vivekanandan Suresh Kumar
PWPhilip Winne
Philip Winne
AHA. Hadwin
A. Hadwin
JNJ. Nesbit
J. Nesbit
DJD. Jamieson-Noel
D. Jamieson-Noel
BSB. Samin
B. Samin
Effects of self-regulated learning (SRL) have been investigated in a variety of contexts. In this research, we explore advanced learning technologies based on the information processing model of SRL in the context of structured programming. We conducted an experiment to study ways to enhance the use of programmers' working memory, to develop tactics to carry out task level activities during programming, and to learn how to program more effectively. The results of the experiment indicate that programmers who received SRL-based treatment outperformed programmers who did not receive the treatment. We argue that the infusion of SRL-based technological interfaces would have a positive influence on the performances in programming.

1 February 2017
Verified
Conference Paper
Closed access
KGKannan Govindarajan
Kannan Govindarajan
VKVivekanandan Suresh Kumar
Vivekanandan Suresh Kumar
Cloud technology is an emerging distributed computing paradigm. It provides Cloud consumers with on-demand computing, storage, and networking resources. A Cloud Resource Broker (CSB) acts as a mediator between cloud consumers and cloud providers. It effectively handles application requests and efficiently allocates cloud resources. Generally, CSB encounters large volumes of application requests. It is essential to properly and evenly distribute these application requests between the available cloud resources. The proposed work introduces a Particle Swarm Optimization (PSO) based load balancing technique, which distributes cloud consumer application requests in an optimal and balanced manner. In addition, the present study investigates the Software-Defined Networking (SDN) cloud infrastructure, which dynamically configures and provides network paths on-demand, based on the network load. The proposed system is simulated and tested on real-world application traces. The proposed PSO based load balancing mechanism is shown to minimize average application response time, while maximizing throughput, cloud consumer satisfaction value, and resource utilization.

1 December 2013
Verified
Conference Paper
Closed access
KGKannan Govindarajan
Kannan Govindarajan
TSThamarai Selvi Somasundaram
Thamarai Selvi Somasundaram
VKVivekanandan Suresh Kumar
Vivekanandan Suresh Kumar
KIKinshuk
Kinshuk
Learners' attainment of academic knowledge in postsecondary institutions is predominantly expressed by summative or formative assessment approaches. Recent advances in educational technology has hinted at a means to measure learning efficiency, in terms of personalization of learner competency and capacity in terms of adaptability of observed practices, using raw data observed from study experiences of learners as individuals and as contributors in social networks. While accurate computational models that embody learning efficiency remain a distant and elusive goal, big data learning analytics approaches this goal by recognizing competency growth of learners, at various levels of granularity, using a combination of continuous, formative and summative assessments. This study discusses a method to continuously capture data from students' learning interactions. Then, it analyzes and clusters the data based on their individual performances in terms of accuracy, efficiency and quality by employing Particle Swarm Optimization (PSO) algorithm.

1 January 2005
Verified
Conference Paper
Closed access
VKVivekanandan Suresh Kumar
Vivekanandan Suresh Kumar
JNJ. Nesbit
J. Nesbit
KHK. Han
K. Han
As differing evaluation instruments are adopted in learning object repositories serving specialized communities of users, what methods can be adopted for translating evaluative data across instruments in order to share this data among different repositories? How can evaluation from different reviewers be properly integrated? How can explicit and implicit measures of preference and quality be combined to recommend objects to users? In this research we studied the application of Bayesian belief network (BBN) to the problem of insufficient and incomplete reviews during learning objects evaluation, and translating and integrating data among different quality evaluation instruments and measures. Two BBNs were constructed to probabilistically model relationships among different roles of reviewers as well as among items of different evaluation measurements. Initial testing using hypothetic data showed that the model was able to make potentially useful inferences about different dimensions of learning object quality. We further extend our model over geographic distances assuming that the reviewers would be distributed and that each reviewer would change the underlying BBN network (to a certain extent) to suit his/her expertise. We highlight issues that arise due to a highly distributed and personalized BBN network that can be used to make valid inferences about learning object quality.