Earticle

e-Learning can be seen as not only one of Internet-based information technologies which can provide education services but also one of teaching-learning methods which can implement self-directed learning. Thus, for evaluation of e-Learning effectiveness, both information-technology-based learning environment and learners' abilities in self-learning and computer-using should be considered simultaneously. This study suggests a research model for evaluating the effectiveness of e-Learning, which is theoretically based on information systems success model, constructivism and self-efficacy. The model is composed of three parts: effectiveness, learning environment, and learners' self-efficacy. Effectiveness is a part of dependent variables: satisfaction and academic performance. Learning environment and learners' self-efficacy can be considered as two sets of explanation variables for effectiveness. The former consists of learning management system, learning contents, and interactions that are provided bye-Learning and the latter means learners' self-regulated efficacy and computer self-efficacy. We show validity of the model empirically by surveying the college students who have experienced e-Learning. In result, most of all hypotheses suggested in this model are accepted in low significant level.

By far, numerous e-government projects requiring huge investments have been conducted in Korea to increase the administrative efficiency and improve the service quality via Internet. However, there's little research focusing on methods and techniques for analyzing and evaluating the projects either to economically justify the investments or to assess the quality of the system and of on-line services to citizens. The purpose of this paper is to suggest a set of factors to take into account for evaluating e-government portals and to empirically test the factors to provide useful implications for building such portals. Based on the literature reviewed, we constructed a research model that includes content, public service, community, design, technology, and portal's attributes as independent variables and the user's satisfaction and the administrative efficiency as dependent variables. A reliability test revealed that the evaluation factors proposed in the paper are sufficiently reliable, and a multi-regression analysis indicated that five hypotheses should be accepted. The findings of the study suggest that emphasis should be placed on public service quality and portal site attributes, among others, when implementing portals.

Technology Acceptance Model (TAM) has been widely used to predict user's behavior to accept the technology. Prior researches have been mainly focused on innovation constructs such as perceived usefulness and perceived ease of use. However, very little research has been conducted to understand individual mental beliefs in technology acceptance and imitation influence. This study integrates Technology Acceptance Model (TAM), Flow Theory (FT) and Diffusion of Innovation Theory (DIT). This paper indicates that imitation context, cognitive absorption (CA) based Flow theory and innovation context are the three important factors influencing user acceptance of information technologies. The proposed model has been tested among 232 users of MP3 players. Results showed that innovation context and cognitive absorption have positive influences on intention to use technology. Not all factors of the imitation context have direct effect on intention to use. However, we found that imitation context has positive influence on intention to use technology through cognitive absorption.

Case-based reasoning (CBR) is a reasoning technique that reuses past cases to find a solution to the new problem. It often shows significant promise for improving effectiveness of complex and unstructured decision making. It has been applied to various problem-solving areas including manufacturing, finance and marketing for the reason. However, the design of appropriate case indexing and retrieval mechanisms to improve the performance of CBR is still a challenging issue. Most of the previous studies on CBR have focused on the similarity function or optimization of case features and their weights. According to some of the prior research, however, finding the optimal k parameter for the k-nearest neighbor (k-NN) is also crucial for improving the performance of the CBR system. In spite of the fact, there have been few attempts to optimize the number of neighbors, especially using artificial intelligence (AI) techniques. In this study, we introduce a genetic algorithm (GA) to optimize the number of neighbors to combine. This study applies the novel approach to Korean stock market. Experimental results show that the GA-optimized k-NN approach outperforms other AI techniques for stock market prediction.

Actual type of aggregation performed by an ordered weighted averaging (OWA) operator heavily depends upon the weighting vector. A number of approaches have been suggested for obtaining the associated weights. In this paper, we present analytic forms of OWA operator weighting functions, each of which has such properties as rank-based weights and constant value of orness, irrespective of number of objectives aggregated. Specifically, we propose four analytic forms of OWA weighting functions that can be positioned at 0.25, 0.334, 0.667, and 0.75 on the orness scale. The merits for using these weights over other weighting schemes can be mentioned in a couple of ways. Firstiy, we can efficiently utilize the analytic forms of weighting functions without solving complicated mathematical programs once the degree of orness is specified a priori by decision maker. Secondly, combined with well-known OWA operator weights such as max, min, and average, any weighting vectors, having a desired value of orness and being independent of the number of objectives, can be generated. This can be accomplished by convex combinations of predetermined weighting functions having constant values of orness. Finally, in terms of a measure of dispersion, newly generated weighting vectors show just a few discrepancies with weights generated by maximum entropy OWA.

Online game business has emerged as the most lucrative entertainment industry, with over 10 million players in South Korea and over 30million in Japan in 2005. While the interactive entertainment market continues to expand, with many new online game publishers entering the market, relatively little theory has been developed about which factors influence online gamers' behavioral intentions (i.e., loyalty, satisfaction, words of mouth, etc.) in this area. The purpose of this research is to investigate the relationships among the gamers' satisfaction, trust toward game publishers, the role of online game community, social reputation, and the managerial support of game publishers. We also examine the differences between Korean and Japanese gamers concerning the relationships of these key success factors. The structural model is tested with the data from entire data samples (i.e., Korean and Japanese gamers pooled together) and each of the sub-samples (i.e., Korean and Japanese gamers taken separately). Properties of the causal paths, including standardized path coefficients, the significance of difference, and variance explained for Trust and Satisfaction in the hypothesized model, are presented. Following the model test, we conduct a test of the differences in path coefficients between Korean and Japanese gamers. Statistical results show that, compared to Japanese gamers, Korean gamers had a greater salient effect on Social Reputation in determining. Trust, in addition to placing a greater emphasis on Support of Game Publishers in determining Social Reputation. Other interesting results concerning game Publishers' strategy are also presented.

The K-means algorithm is widely used at the initial stage of data analysis in data mining process, partly because of its low time complexity and the simplicity of practical implementation. Cluster validity indices are used along with the algorithm in order to determine the number of clusters as well as the clustering results of datasets. In this paper, we present a performance comparison of sixteen indices, which are selected from forty indices in literature, while considering their applicability to nonhierarchical clustering algorithms. Data sets used in the experiment are generated based on multivariate normal distribution. In particular, four error types including standardization, outlier generation, error perturbation, and noise dimension addition are considered in the comparison. Through the experiment the effects of varying number of points, attributes, and clusters on the performance are analyzed. The result of the simulation experiment shows that Calinski and Harabasz index performs the best through the all datasets and that Davis and Bouldin index becomes a strong competitor as the number of points increases in dataset.