There is currently a shortage of graduates from Computer Science and Information Technology programs to meet industry demands, yet these programs have a difficult time recruiting and retaining students. Attrition in some programs range from 50 to 85% of those students who first enter. Prior studies indicate that personality attributes, learning styles, time management skills and logical thinking abilities may all be valid predictors of those students who are likely to be successful in studying computer science. The present study used a first year computer programming class to collect data on these variables at the start of the semester and then compared them with the students’ academic result at the conclusion of the course. The study found that personality attributes, learning style, and time management skills were poor predictors of future success in computer programming. However, certain groups of logical problems, particularly those requiring cognitive modeling of the problem, disjunctive logic and deductive and rule based reasoning can serve as future predictors of success in this field. These results can be used to create an admissions test for students considering studying computer science and suggest modifications to educational curricula within the primary and secondary school systems.
The Great Plains Free-Net (GPFN), a community-owned and operated Internet Service Provider in Regina, Saskatchewan, was embarking on a comprehensive modernization strategy. As part of this strategy, GPFN wanted to offer a number of new services and adopting a more member-centric philosophy. One problem with the previous User Management system was that very few volunteers understood or could manage it. It relied on an inter-related set of some 300 shell scripts and flat file tables from which to manage its database. The new design utilized a relational database engine for the back-end and a secure, web-enabled interface from which members, users, guests and volunteers could apply for accounts, upgrade accounts and approve, reject or revoke applications and accounts.
The system was designed to use a volunteer pool of developers to develop and implement the new system using a combination of MySQL, PERL, HTML and CSS as these were the skills the organization had in house.
Merchants are sensitive to costs. An effective e-commerce model will provide a lower cost to merchants. These costs include the development costs and licensing and setup fees to payment enable their site as well as the ongoing costs of maintaining their site and any per-month or per-transaction fees. For these reasons, off-the-shelf e-commerce software, which often involves large initial software costs and ongoing licensing fees, are difficult to justify.
A survey iwas conducted to assist Great Plains Free-Net in better understanding the needs of the community as well as helping Great Plains Free-Net understand how to better communicate its concept to the community as a whole and helping recruit prospective information providers and determine interests among the community. These surveys found a strong interest in Great Plains Free-Net as an information and connection service. Given the demonstrated community support for the Free-Net, Great Plains Free-Net should have no difficulty justifying its existence to users, information providers and funding agencies.
Bayesian Networks are becoming an increasingly important area for research and application in the entire field of Artificial Intelligence. This paper explores the nature and implications for Bayesian Networks beginning with an overview and comparison of inferential statistics and Bayes' Theorem. The nature, relevance and applicability of Bayesian Network theory for issues of advanced computability forms the core of the current discussion. A number of current applications using Bayesian networks is examined. The paper concludes with a brief discussion of the appropriateness and limitations of Bayesian Networks for human-computer interaction and automated learning.
This paper challenges the currently accepted disease model of alcoholism as inaccurate and ineffective. In its place, a paradigm based on operant and classical conditioning learning models is examined and proposed as a more useful and pragmatic approach to understanding this problem.