Large Computer Science Research Articles

Practical Sentiment Analysis for Education: The Power of Student Crowdsourcing

Sentiment analysis provides a promising tool to automatically assess the emotions voiced in written student feedback such as periodically collected unit-of-study reflections. The commonly used dictionary-based approaches are limited to major languages and fail to capture contextual differences. Pretrained large language models have been shown to be biased and online versions raise privacy concerns. Hence, we resort to traditional supervised machine learning (ML) approaches which are designed to overcome these issues by learning from domain-specific labeled data. However, these labels are hard to come by -- in our case manually annotating student feedback is prone to bias and time-consuming, especially in high-enrollment courses. In this work, we investigate the use of student crowdsourced labels for supervised sentiment analysis for education. Specifically, we compare crowdsourced and student self-reported labels with human expert annotations and use them in various ML approaches to evaluate the performance on predicting emotions of written student feedback collected from large computer science classes. We find that the random forest model trained with student-crowdsourced labels tremendously improves the identification of reflections with negative sentiment. In addition to our quantitative study, we describe our crowdsourcing experiment which was intentionally designed to be an educational activity in an introduction to data science course.

Motivational climate predicts effort and achievement in a large computer science course: examining differences across sexes, races/ethnicities, and academic majors

BackgroundThe motivational climate within a course has been shown to be an important predictor of students’ engagement and course ratings. Because little is known about how students’ perceptions of the motivational climate in a computer science (CS) course vary by sex, race/ethnicity, and academic major, we investigated these questions: (1) To what extent do students’ achievement and perceptions of motivational climate, cost, ease, and effort vary by sex, race/ethnicity, or major? and (2) To what extent do the relationships between students’ achievement and perceptions of motivational climate, cost, and effort vary by sex, race/ethnicity, and major? Participants were enrolled in a large CS course at a large public university in the southeastern U.S. A survey was administered to 981 students in the course over three years. Path analyses and one-way MANOVAs and ANOVAs were conducted to examine differences between groups.ResultsStudents’ perceptions of empowerment, usefulness, interest, and caring were similar across sexes and races/ethnicities. However, women and Asian students reported lower success expectancies. Students in the same academic major as the course topic (i.e., CS) generally reported higher perceptions of the motivational climate than students who did not major or minor in the course topic. Final grades in the course did not vary by sex or race/ethnicity, except that the White and Asian students obtained higher grades than the Black students. Across sex, race/ethnicity, and major, students’ perceptions of the motivational climate were positively related to effort, which was positively related to achievement.ConclusionsOne implication is that females, Asian students, and non-CS students may need more support, or different types of support, to help them believe that they can succeed in computer science courses. On average, these students were less confident in their abilities to succeed in the course and were more likely to report that they did not have the time needed to do well in the course. A second implication for instructors is that it may be possible to increase students’ effort and achievement by increasing students’ perceptions of the five key constructs in the MUSIC Model of Motivation: eMpowerment, Usefulness, Success, Interest, and Caring.

Interactions between self‐regulated learning and assessment for learning in an undergraduate introductory computer science course

AbstractMany introductory computer science (CS) courses are intended to address the increased demand for computer literacy and the development of cross‐cutting concepts and practices of computational thinking (CT). Colleges and universities offer introductory CS courses every semester toward this end. The issue is centered on how to support CT learning in a large introductory CS course. To do it, an introductory CS course was designed to enrich CT by embedding short episodes of assessments to facilitate learning (AfL) and to support students’ self‐regulated learning (SRL). This chapter begins by articulating the intricacies of the underlying processes of SRL, principles of AfL, and computational thinking practices in CS. Using the undergraduate introductory CS course referred to above, key practices of SRL and AfL will be illustrated to demonstrate how student learning of CS programming is structured with various online and offline supports. The chapter then concludes by appraising aspects of this CS course with practical suggestions for supporting student SRL and their learning of CS programming.

Scaling anti‐plagiarism efforts to meet the needs of large online computer science classes: Challenges, solutions, and recommendations

AbstractBackgroundPlagiarism is a very serious offence in academic institutions. Yet there is some reluctance to address plagiarism by educators as its enforcement can require a significant time commitment if not handled wisely. Handling plagiarism at scale has the potential to exacerbate this problem.ObjectivesThis article explores the challenges educators face when it comes to enforcing plagiarism from within the course environment, presents a solution that addresses these challenges, and provides guidelines for preventing and detecting plagiarism in our at‐scale courses.MethodsA workflow is shared which was created from our experience of handling plagiarism within our large online computer science graduate level program. We present an empirical study of this workflow that shows the overall prevalence of misconduct.Results and ConclusionsResults demonstrate this workflow being effectively and efficiently applied to address plagiarism across four courses and three semesters by a single individual working half time. We have observed that warnings on low‐stakes assignments can be very effective at deterring future misconduct. We have also observed that cases come to a quicker resolution when students readily acknowledge their own misconduct after being promptly notified.Takeaways from the studyPlagiarism should not only be managed through institution wide policies, but also through effective strategies implemented in the course environment. This is particularly important when handling plagiarism at scale. Addressing plagiarism starts in the course and therefore effective strategies for handling it there are needed.

An ensembling model for early identification of at‐risk students in higher education

AbstractAudience Response Systems like clickers are gaining much attention for early identification of at‐risk students as quality education, student success rate and retention are major concerning areas, as evidenced in this COVID scenario. Usage of this active learning strategy across the varying strength of classrooms are found to be much effective in retaining the attention, retention and learning power of the students. However, implementing clickers for large classrooms incur overhead costs on instructor's part. As a result, educational researchers are experimenting with various lightweight alternatives. This paper discusses one such alternative: lightweight formative assessments for blended learning environments. It discusses their implementation and effectiveness in early identification of at‐risk students. This study validates the usage of lightweight assessments for three core pedagogically different courses of large computer science engineering classrooms. It uses voting ensemble classifier for effective predictions. With the usage of lightweight assessments in early identification of at‐risk students, accuracy range of 87%–94.7% have been achieved along‐with high ROC‐AUC values. The study also proposes the generalized pedagogical architecture for fitting in these lightweight assessments within the course curriculum of pedagogically different courses. With the constructive outcomes, the light‐weight assessments seem to be promising for efficient handling of scaling technical classrooms.

Log Analysis-Based Resource and Execution Time Improvement in HPC: A Case Study

High-performance computing (HPC) uses many distributed computing resources to solve large computational science problems through parallel computation. Such an approach can reduce overall job execution time and increase the capacity of solving large-scale and complex problems. In the supercomputer, the job scheduler, the HPC’s flagship tool, is responsible for distributing and managing the resources of large systems. In this paper, we analyze the execution log of the job scheduler for a certain period of time and propose an optimization approach to reduce the idle time of jobs. In our experiment, it has been found that the main root cause of delayed job is highly related to resource waiting. The execution time of the entire job is affected and significantly delayed due to the increase in idle resources that must be ready when submitting the large-scale job. The backfilling algorithm can optimize the inefficiency of these idle resources and help to reduce the execution time of the job. Therefore, we propose the backfilling algorithm, which can be applied to the supercomputer. This experimental result shows that the overall execution time is reduced.

A Review of Cost and Makespan-Aware Workflow Scheduling in Clouds

Scientific workflow is a common model to organize large scientific computations. It borrows the concept of workflow in business activities to manage the complicated processes in scientific computing automatically or semi-automatically. The workflow scheduling, which maps tasks in workflows to parallel computing resources, has been extensively studied over years. In recent years, with the rise of cloud computing as a new large-scale distributed computing model, it is of great significance to study workflow scheduling problem in the cloud. Compared with traditional distributed computing platforms, cloud platforms have unique characteristics such as the self-service resource management model and the pay-as-you-go billing model. Therefore, the workflow scheduling in cloud needs to be reconsidered. When scheduling workflows in clouds, the monetary cost and the makespan of the workflow executions are concerned with both the cloud service providers (CSPs) and the customers. In this paper, we study a series of cost-and-time-aware workflow scheduling algorithms in cloud environments, which aims to provide researchers with a choice of appropriate cloud workflow scheduling approaches in various scenarios. We conducted a broad review of different cloud workflow scheduling algorithms and categorized them based on their optimization objectives and constraints. Also, we discuss the possible future research direction of the clouds workflow scheduling.

Weighted and deflated global GMRES algorithms for solving large Sylvester matrix equations

The solution of a large-scale Sylvester matrix equation plays an important role in control and large scientific computations. In this paper, we are interested in the large Sylvester matrix equation with large dimensionA and small dimension B, and a popular approach is to use the global Krylov subspace method. In this paper, we propose three new algorithms for this problem. We first consider the global GMRES algorithm with weighting strategy, which can be viewed as a precondition method. We present three new schemes to update the weighting matrix during iterations. Due to the growth of memory requirements and computational cost, it is necessary to restart the algorithm effectively. The deflation strategy is efficient for the solution of large linear systems and large eigenvalue problems; to the best of our knowledge, little work is done on applying deflation to the (weighted) global GMRES algorithm for large Sylvester matrix equations. We then consider how to combine the weighting strategy with deflated restarting, and propose a weighted global GMRES algorithm with deflation for solving large Sylvester matrix equations. In particular, we are interested in the global GMRES algorithm with deflation, which can be viewed as a special case when the weighted matrix is chosen as the identity. Theoretical analysis is given to show rationality of the new algorithms. Numerical experiments illustrate the numerical behavior of the proposed algorithms.

Accelerating Dense Matrix Computations with Effective Workload Partitioning on Heterogeneous Architectures

ABSTRACTThe emergence of High Performance Computing (HPC) has enabled the researchers to perform large scientific computations efficiently and quickly. But as the heterogeneity of the processing units of the HPC systems increased, the utilization of all the resources became an issue. Fully harnessing the power of these systems requires efficient division of work across all the processing units. This solves the issue of under-utilization of resources and improves performance of the application. In this research work, we present a dynamic approach to workload partitioning that obtains the optimal workload partition and schedules them to processing units for parallel processing. Our workload partitioning technique is able to respond automatically to performance variation to provide good performance, it requires very negligible training and is implemented as a library. Performance results show that our dynamic approach is better than static and linear approach. By running the Dense Matrix-Matrix Multiplication kernel library by our proposed method on both CPU and Graphics Processing Unit (GPU) in parallel, we obtain average speedups from to over CPU and to over GPU. We used our method on multi-GPUs for which we obtain average speedups of over CPU and over single GPU.

Class Size and Confidence Levels Among Female STEM Students [Impact

An encouraging aspect of the results from this study is that they are actionable. Ideally, more schools will work to reduce class sizes, particularly in large introductory computer science and STEM courses. If that's not an option, labs and discussion sections should be small to create an environment where students feel comfortable engaging. Harvey Mudd, which measured little gender-based confidence gap in our initial study, offers different introductory computer sciences classes based on experience. Students new to computer science do not take the same introductory classes as (mostly male) students with years of coding experience. Technology can also help by enabling anonymity. Research by Jong et al. found that anonymity increased classroom participation rates and improved learning performance. They wrote that “by using a system that helps group members conceal their identities, students are less affected by interpersonal relationships and peer pressure and are therefore more willing to participate fully in discussions, learning more from the process as a result”. There are numerous issues contributing to the gender gap in technology fields: women graduate in fewer numbers with computer science and engineering degrees, are hired in fewer numbers into technical roles by leading technology firms, and are less likely to stay in their fields. Research like this lends a window into specific actions school instructors and administrators can take to begin minimizing that gap. Reducing class sizes is a start, but should not be the end.

Understanding and modeling diverse scientific careers of researchers

a b s t r a c t This paper analyzes the diverse scientific careers of researchers in order to understand the key factors that could lead to a successful career. Essentially, we intend to answer some specific questions pertaining to a researcher's scientific career - What are the local and the global dynamics regulating a researcher's decision to select a new field of research at dif- ferent points of her entire career? What are the suitable quantitative indicators to measure the diversity of a researcher's scientific career? We propose two entropy-based metrics to measure a researcher's choice of research topics. Experiments with large computer science bibliographic dataset reveal that there is a strong correlation between the diversity of the career of a researcher and her success in scientific research in terms of the number of cita- tions. We observe that while most of the researchers are biased toward either adopting diverse research fields or concentrating on very few fields, a majority of the prominent researchers tend to follow a typical scatter-gather policy - although their entire careers are immensely diverse with different types of fields selected at different time periods, they remain focused primarily in at most one or two fields at any particular time point of their career. Finally, we propose a stochastic model which, quite accurately, mimics the notion of field selection process observed in the real publication dataset.

Rubyric

Writing feedback can be very time-consuming on large Computer Science courses. Evolving from years of research on computer-assisted grading tools, we present the Rubyric assessment tool which allows graders to effortlessly construct feedback with reusable phrases. The tool can be used to facilitate the manual assessment of assignments varying from essays to programming projects.

Reducing lecture and increasing student activity in large computer science courses

This paper describes how one instructor uses a web-based tool called MessageGrid in a laptop-enhanced computer science course to accomplish five goals: (1) to solicit questions from students regarding pre-lecture reading assignments, (2) to engage the students in classroom activity that supports the lecture material for the day, (3) to conduct short, frequent assessments of student comprehension, (4) to conduct exercises that encourage peer-learning, and (5) to use Ink -based software animations to supplement the lecture material for the day. We present MessageGrid and explain how the short lecture exercises (called lexes ) help engage the students in class. We show how the assessment component of MessageGrid helps students know whether or not they learning the material as they should, and enables the instructor to identify students who are falling behind. We present four different ways of designing lexes. We end with conclusions and plans for future work.

Tools and techniques for large scale grading using Web-based commercial off-the-shelf software

Courseware/Course Management Systems (CMS) such as WebCT or Blackboard are an increasingly popular way to provide a web presence for a course. However, their current web-browser reliance makes it difficult for them to provide functionality that could be useful to computer science instructors. This paper describes our augmentation of a CMS in a large introductory computer science class. It further describes our enhancement of the CMS by clientside software (i.e. residing on the graders computer), written for use by the instructors and graders. Finally, it indicates how conventional CMS architecture can be extended to provide additional functionality that would be desirable for computer science instruction.

Using undergraduates as teaching assistants at a state university

This paper describes a program that uses undergraduates as teaching assistants to staff large computer science classes, particularly at the introductory level. Creating such a program at a state school presented special challenges, but the program has become a mainstay for the department's undergraduate program. The program has been so successful that we have expanded it to cover some sophomore and junior level courses, including a discrete mathematics course. Among the benefits of the program are reduced overall cost, improved quality of instruction, the formation of an undergraduate community and the practical experience that the undergraduate teaching assistants themselves gain.

Management challenges in a large introductory computer science course

Many colleges and universities throughout the world offer introductory computer science courses with significant student enrollment. There are certainly plenty of challenges for the administrators of those courses; ensuring equitable grading across different instructors and offering times for the course, accomplishing the massive material preparation and other logistical tasks required for the course, grading the large number of assessments associated with the large enrollment, and managing the numerous instructors in the course are a few examples. This paper discusses the processes we have implemented to address these and other management challenges at the U.S. Air Force Academy.

Master–Worker: An Enabling Framework for Applications on the Computational Grid

The goal of this work is to create a tool that allows users to easily distribute large scientific computations on computational grids. Our tool MW relies on the simple master–worker paradigm. MW provides both a top Level interface to application software and a bottom Level interface to existing Grid computing toolkits. Both interfaces are briefly described. We conclude with a case study, where the necessary Grid services are provided by the Condor high-throughput computing system, and the MW-enabled application code is used to solve a combinatorial optimization problem of unprecedented complexity.

Object-Oriented Techniques in Large Scientific Computing Projects: Experience with C++

First Page

SIMULTANEOUS PARALLEL REDUCTION ON SIMD MACHINES

Proper distribution of operations among parallel processors in a large scientific computation executed on a distributed-memory machine can significantly reduce the total computation time. In this paper, we propose an operation called simultaneous parallel reduction(SPR), that is amenable to such optimization. SPR performs reduction operations in parallel, each operation reducing a one-dimensional consecutive section of a distributed array. Each element of the distributed array is used as an operand to many reductions executed concurrently over the overlapping array's sections. SPR is distinct from a more commonly considered parallel reduction which concurrently evaluates a single reduction. In this paper we consider SPR on Single Instruction Multiple Data (SIMD) machines with different interconnection networks. We focus on SPR over sections whose size is not a power of 2 with the result shifted relative to the arguments. Several algorithms achieving some of the lower bounds on SPR complexity are presented under various assumptions about the properties of the binary operator of the reduction and of the communication cost of the target architectures.

Peer learning in an introductory computer science course

A problem in teaching large introductory computer science courses is to overcome the impersonality of the large lecture class and to provide more personal attention to individual students. Our approach is to use peer learning experiences to instill in students the need to take responsibility for their learning and for the learning of those around them. Recent work has shown that educational quality for students and productivity for faculty can be enhanced through use of peer-learning environments where students do not just learn and faculty do not just teach. The novel aspects of our work are to apply group learning in a large introductory computer science class setting and to expect more responsibility on the part of students for their learning. In support of these goals we have introduced the use of upper-level undergraduate students to help facilitate student group interaction. In addition, we have developed software to minimize the administrative overhead of handling many groups and for students to electronically record group learning activities.