Number Of Computer Science Research Articles

Technical Note—On Adaptivity in Nonstationary Stochastic Optimization with Bandit Feedback

Optimal Nonstationary Optimization Without Knowing Function Changes Nonstationary stochastic optimization plays a vital role in a number of computer science and operations research applications. It is known how to design and analyze algorithms that optimize a sequence of strongly convex/concave and smooth functions with access to only one-point noisy function values with the underlying function sequence subject to maximum magnitude of function changes. In recent work from Wang titled “Technical Note: On Adaptivity in Nonstationary Stochastic Optimization with Bandit Feedback,” an optimization algorithm is designed and analyzed without assuming the magnitude of function changes is known in advance. Optimality of the designed algorithm is demonstrated.

The Association of High School Computer Science Content and Pedagogy with Students’ Success in College Computer Science

The number of computer science (CS) courses has been dramatically expanding in U.S. high schools (HS). In comparison with well-established courses in mathematics and science, little is known about how the decisions made by HS CS teachers regarding how and what to teach impact student performance later in introductory college CS courses. Drawing on a large sample of 2,871 introductory college CS students at 115 U.S. institutions who had taken a CS course in HS, we examined the topic coverage and prevailing instructional methods in the HS course and investigated how these experiences influenced student performance in college CS. Controlling for differences in student background, we find two predictors of higher grades in college CS: greater frequency of coding-related activities in HS (programming, debugging, studying algorithms) and lower frequency of “non-coding” computer use (e.g., data analysis, computer security). Interaction models revealed a more complex story. Coding-related activity more heavily benefited students who did not have coding help available at home. In the 28% of college CS courses in which instructors employed innovative pedagogies, students with higher ACT or SAT mathematics scores had a greater advantage than in traditionally taught courses. Finally, in the innovative college courses, students whose HS CS exams had typically included testing on vocabulary did worse than students whose exams had not included such tests.

Dot.com Boom and Bust Effects on MIS College Enrollments: 1995 – 2006

ABSTRACT This paper presents an overview of the impact of the dot.com boom and bust on the enrollment of students in information technology related business school programs at seven US colleges and universities during the period 1995-2006. Also, this paper presents the results of a related survey of factors affecting undergraduate business students' choice of major at two southern US universities. INTRODUCTION Many are familiar with the dot.com boom at the end of the 20th century and the subsequent bust in the first few years of the 21st century. Investors became overnight millionaires as Internet stock prices soared, and hundreds of companies went bankrupt in the subsequent decline. The beginning of the e-commerce boom began around 1996 when Internet stock tracking indices first appeared. As seen in Figure 1, the peak in stock values for most e-commerce businesses occurred in the early part of 2000, while the end of the decline occurred in late 2002. Since then, prices have recovered somewhat, but nearly 50% of the businesses operating in 1998 no longer existed in 2003 (University of Maryland, 2006). In addition to the wild fluctuation of stock prices, many other related activities were affected such as advertising expenditures, real estate and rental prices in Silicon Valley, and even the number of academic papers written about electronic commerce (Kim, Aiken, & Vanjani, 2006). Also, while salaries and hiring of information technology (IT) professionals were affected by this bubble to some extent, the enrollments of IT students in colleges and universities were affected to a much larger degree. The purpose of this paper is to analyze the relationship between the dot.com boom and bust with IT college enrollments between the years of 1995 and 2006 and investigate factors affecting the related enrollment decline. This paper presents a summary of IT related major enrollment at seven different universities in the United States from 1995 through 2006. Also, a survey was used to identify the factors that affected the choice of major for undergraduate business students. The results of this survey of factors affecting undergraduate business students' choice of majors conducted at two universities shows that personal interest in and reputation of the major were significant factors in students' choice of major. It is the expectation that this information will likely better prepare colleges and universities to plan recruitment strategies for IT majors. BACKGROUND The history of IT student enrollments at universities and colleges in the United States has shown erratic periods of growth (Lennox, Woratschek, & Davis, 2005; Lomerson & Pollacia, 2006). For example, from 1980 to 1986, the number of Computer Science (CS) students increased by more than four times, but between 1986 and 1990, their number had dropped 40% (Cale, Mawhinney, & Callaghan, 1991; Office of Technology Policy, 1997). By the early 1990s, enrollments again began to grow. However, it was the beginning of the dot.com bubble that really began to spur enrollment growth. For example, at the University of Mississippi, growth in the number of Management Information Systems (MIS) majors had increased by roughly 50% over the first half of the decade (see Figure 2), and many other universities experienced similar bubbles in IT enrollment growth. As an additional example, Florida State's MIS student population doubled between 1995 and 2000 before declining substantially (see Figure 3). [FIGURE 1 OMITTED] [FIGURE 2 OMITTED] [FIGURE 3 OMITTED] IT enrollment at universities across the country increased dramatically at the end of the decade as many students became aware of the great job offers and the high salaries in the field because of the boom in e-commerce hiring (Goff, 2000; Goff, 2001) and to a lesser extent, the need for programmers to fix the Y2K bug (Edwards, 1999). …

Personality types and Facebook advertising: An exploratory study

The Five-Factor Model of personality types has been utilized by a number of computer science and psychology academics as a framework to explore aspects of internet adoption and, more recently, social media adoption and behaviours. While the use of personality type models in research has been less popular with marketing academics, the Five-Factor Model has been used to predict the emotional appeal of advertising based on personality types and potentially to provide guidelines for tailoring adverts accordingly. Given the growing impact of advertising on social media platforms, especially Facebook’s ongoing refinements to its advertising models, the question arises for both academics and practitioners alike as to whether personality type models such as Five Factor can successfully be used to predict consumer responses to promotional messages delivered via social media. This paper describes an exploratory study in which participants were shown a series of Facebook promotional posts and asked to describe their responses, and to complete a questionnaire to assess Five-Factor personality type. The results of the study suggest that responses to Facebook promotional messages may indeed vary based on the user’s personality type, but further research is needed to validate these findings. The study also underscores the challenges of assessing personality type in time-limited studies and suggests that further research is needed to evaluate the suitability of this approach for practitioners.