South African Computer Journal - Special issue 1, August 2014
Volumes & issues
Special issue 1, August 2014
Source: South African Computer Journal 53 (2014)More Less
In this special issue, we feature selected papers from the 2013 South African Institute of Computer Scientists and Information Technologists Research (SAICSIT 2013) conference, hosted by Rhodes University at the Blue Lagoon Hotel in East London, 7-9 October. The authors of seven of the best SAICSIT papers were invited to submit extended versions of their research for this Special Issue. A second round of reviewing resulted in the four papers published here; two from Computer Science and two from Information Systems.
Source: South African Computer Journal 53, pp 1 –14 (2014)More Less
In indexing of, and pattern matching on, DNA and text sequences, it is often important to represent all factors of a sequence. One efficient, compact representation is the factor oracle (FO). At the same time, any classical deterministic finite automaton (DFA) can be transformed to a so-called failure one (FDFA), which may use failure transitions to replace multiple symbol transitions, potentially yielding a more compact representation. We combine the two ideas and directly construct a failure factor oracle (FFO) from a given sequence, in contrast to ex post facto transformation to an FDFA. The algorithm is suitable for both short and long sequences. We empirically compared the resulting FFOs and FOs on number of transitions for many DNA sequences of lengths 4 - 512, showing gains of up to 10% in total number of transitions, with failure transitions also taking up less space than symbol transitions. The resulting FFOs can be used for indexing, as well as in a variant of the FO-using backward oracle matching algorithm. We discuss and classify this pattern matching algorithm in terms of the keyword pattern matching taxonomies of Watson, Cleophas and Zwaan. We also empirically compared the use of FOs and FFOs in such backward reading pattern matching algorithms, using both DNA and natural language (English) data sets. The results indicate that the decrease in pattern matching performance of an algorithm using an FFO instead of an FO may outweigh the gain in representation space by using an FFO instead of an FO.
The expatriate information flow model : towards understanding Internet usage in the Kingdom of Saudi Arabia : research articleSource: South African Computer Journal 53, pp 15 –31 (2014)More Less
Expatriate adjustment research has identified a number of challenges that expatriates experience when adjusting to the host country. These include spousal influence, cultural training/understanding, fluency in the host language and the personality or emotional readiness of the expatriate. These challenges are amplified when considered in the Kingdom of Saudi Arabia (KSA), which has large cultural distance when compared to the average Western culture and therefore provides a setting for an interesting study. This paper describes how the "degree of information flow", a substantive category derived through the grounded theory methodology, provides an understanding of the emotional relationship expatriates in KSA have with the Internet. An expatriate information flow model was developed explaining the pre-conditions to and the consequences of information flow. The paper shows that the pre-conditions to information flow include overcoming the challenges experienced by Internet users in KSA in addition to their intention to use the Internet based on their personal needs, status and personality. The consequences of information flow were indicated as having an overall positive impact on the expatriate's well-being. This was operationalised as giving expatriates an extended control over their environment, increased social presence and an increased exchange of information between the expatriates and their benevolent communities.
Usability evaluation for Business Intelligence applications : a user support perspective : research articleSource: South African Computer Journal 53, pp 32 –44 (2014)More Less
Business Intelligence (BI) applications provide business information to drive decision support. Usability is one of the factors determining the optimal use and eventual benefit derived from BI applications. The documented need for more BI usability research together with the practical necessity for BI evaluation guidelines in the mining industry provides the rationale for this study. The purpose of the study was to investigate the usability evaluation of BI applications in the context of a coal mining organization. The research is guided by the question: How can the existing usability criteria be customized to evaluate the usability of BI applications? The research design included user observation, heuristic evaluation and a survey. Based on observations made during user support on a BI application used at a coal mining organization a log of usability issues was compiled. The usability issues extracted from this log were compared and contrasted with general usability criteria from literature to synthesize an initial set of BI usability evaluation criteria. These criteria were used as the basis for a heuristic evaluation of the BI application used at the coal mining organization. The same BI application was also evaluated using the Software Usability Measurement Inventory (SUMI) standardized questionnaire. The results from the two evaluations were triangulated and then compared with the BI user issues again to contextualize the findings and synthesize a validated and refined set of criteria. The main contribution of the study is the usability evaluation criteria for BI applications presented as guidelines. These BI guidelines deviate from existing usability evaluation guidelines in that it emphasises the aspects of information architecture, learnability and operability.
Source: South African Computer Journal 53, pp 45 –59 (2014)More Less
General-purpose computation on graphics processing units (GPGPU) has great potential to accelerate many scientific models and algorithms. However, since some problems are considerably more difficult to accelerate than others, ascertaining the effort required to accelerate a particular problem is challenging. Through the acceleration of three typical scientific problems, seven problem attributes have been identified to assist in the evaluation of the difficulty of accelerating a problem on a GPU. These attributes are inherent parallelism, branch divergence, problem size, required computational parallelism, memory access pattern regularity, data transfer overhead, and thread cooperation. Using these attributes as difficulty indicators, an initial problem difficulty classification framework has been created that aids in evaluating GPU acceleration difficulty. The difficulty estimates obtained by applying the classification framework to the three case studies correlate well with the actual effort expended in accelerating each problem.