Zaiyong Tang

The Demise of Novell Netware: Did Perceptions Related to Network Administration Play a Role?

In this study the authors examine some possible reasons for Novell Netwares fall from dominance in the 1990s to its near disappearance. The authors examine the role of the external variables security, productivity, and sharing on an administrators perceptions and intention to use the software within the Technology Acceptance Model (TAM). The results show that difficulty in administration was not responsible for Netwares demise; on the contrary, Netwares environment resulted in a positive intention to use it.

Globally Convergent Particle Swarm Optimization via Branch-and-Bound

Particle swarm optimization (PSO) is a recently developed optimization method that has attracted interest of researchers in various areas. PSO has been shown to be effective in solving a variety of complex optimization problems. With properly chosen parameters, PSO can converge to local optima. However, conventional PSO does not have global convergence. Empirical evidences indicate that the PSO algorithm may fail to reach global optimal solutions for complex problems. We propose to combine the branch-and-bound framework with the particle swarm optimization algorithm. With this integrated approach, convergence to global optimal solutions is theoretically guaranteed. We have developed and implemented the BB-PSO algorithm that combines the efficiency of PSO and effectiveness of the branch-and-bound method. The BB-PSO method was tested with a set of standard benchmark optimization problems. Experimental results confirm that BB-PSO is effective in finding global optimal solutions to problems that may cause difficulties for the PSO algorithm.

Global IT Expenditure Growth: An Empirical Investigation Across Some Developing Nations

Information technology (IT) expenditures in different developing nations of the world have been impressive but controversial lately. Research is needed to know how IT expenditures are growing in these nations. Do stages of IT development or price drops in IT infrastructure influence such growth? We intend to explore these issues with various growth models, using data from 14 nations (constrained by data availability) over a period of time. Our preliminary results show that previous IT expenditure growth models can be improved by including the impact of price and that a price drop in IT keeps the growth unabated. Preliminary evidence also suggests that developing nations are benefiting from a price drop.

Explorative Assessment of Internet Hacking: An Agent-Based Modeling Approach

Abstract Internet hacking is fast becoming a significant threat not only to businesses, but government entities, online communities, and individual Internet users as well. We have built an agent-based model (ABM) to study the dynamics of Internet hacking. Several factors that impact the adoption of Internet hacking are evaluated. Through ABM simulations we explore the interactions of various types of Internet users along with their hacking propensity and the resulting hacking trends. The simulations point to several interesting outcomes. For instance, the hacking trend is greatly affected by the quantum of law enforcement and by the influence of hackers on normal users. On the other hand, the number of initial hackers and the degree of interaction do not appear to be significant factors. In addition, the results of the simulation illustrate the impact of the mass media and of “hacking websites” on Internet hacking trends.

Improving population estimation with neural network models

Intercensal and postcensal population estimates are essential in federal, state, and local governments planning and resource allocation. Traditionally, linear regression based models are widely used for projecting population distributions in a given region. We constructed population projection models with various types of artificial neural networks. Using historical census data, we tested the performance of the neural network models against the ratio correlation regression model that we have used for the last 20 years. The results indicate that properly trained neural networks outperform the regression model in both model fitting and projection. Among the different neural network models we tested, the fuzzy logic based neural network performed the best.

Pruning feedforward neural network search space using local lipschitz constants

Combination of backpropagation with global search algorithms such as genetic algorithm (GA) and particle swarm optimization (PSO) has been deployed to improve the efficacy of neural network training. However, those global algorithms suffer the curse of dimensionality. We propose a new approach that focuses on the topology of the solution space. Our method prunes the search space by using the Lipschitzian property of the criterion function. We have developed procedures that efficiently compute local Lipschitz constants over subsets of the weight space. Those Local Lipschitz constants can be used to compute lower bounds on the optimal solution.

The use of a flexible diffusion model for forecasting national-level mobile telephone and internet diffusion

Much attention has been given to adoption and diffusion, defined as the degree of market penetration, of Information and Communications Technologies (ICT) in recent years (Carter, Jambulingam, Gupta, & Melone, 2001; Kiiski & Pohjola, 2002; Milner, 2003; Benhabib & Spiegel, 2005). The theory of diffusion of innovations considers how a new idea spreads throughout the market over time. The ability to accurately predict new product diffusion is of concern to designers, marketers, managers, and researchers alike. However, although the diffusion process of new products is generally accepted as following an s-curve pattern, where diffusion starts slowly, grows exponentially, peaks, and then declines (as shown in Fig. 1), there is considerable disagreement about what factors affect diffusion and how to measure diffusion rates (Bagchi, Kirs, & Lopez, 2008).