An extended tuning method for cost-sensitive regression and forecasting

In many real-world regression and forecasting problems, over-prediction and under-prediction errors have different consequences and incur asymmetric costs. Such problems entail the use of cost-sensitive learning, which attempts to minimize the expected misprediction cost, rather than minimize a simple measure such as mean squared error. A method has been proposed recently for tuning a regular regression model post hoc so as to minimize the average misprediction cost under an asymmetric cost structure. In this paper, we build upon that method and propose an extended tuning method for cost-sensitive regression. The previous method becomes a special case of the method we propose. We apply the proposed method to loan charge-off forecasting, a cost-sensitive regression problem that has had a bearing on bank failures over the last few years. Empirical evaluation in the loan charge-off forecasting domain demonstrates that the method we have proposed can further lower the misprediction cost significantly.     

Impact of network effects and diffusion channels on home computer adoption

Researchers have identified network effects as one of the major drivers for the adoption and diffusion of household technologies. However, the mechanisms and social contexts through which network effects induce technology adoption are unclear. In this paper, we investigate the adoption of household computers using the dataset from the 1989-2003 Computer and Internet Supplement to the Current Population Surveys (CPS). We argue that social influence and learning play a dominant role in inducing the adoption of home computers, and workplaces and schools are important channels through which network effects take place. We find that recent adopters have a stronger impact than distant adopters on future computer adoptions. When the adoption rate is low, channels play a more important role, but their effects diminish as the adoption rate grows. We also find that diffusion channels are more effective for first-time adoption than for repeat purchases. Overall, our study provides important theoretical, policy, and managerial implications.     

Pricing strategies for tied digital contents and devices

Media companies are increasingly offering digital content to consumers. Many of these companies are tying digital content with their proprietary digital devices. In this study, we develop a consumer demand model for digital device and digital content based on a constant elasticity demand function. In modeling consumer valuation of the digital device, we take consumer surplus on digital content into account. We further derive equilibrium prices for digital devices based on an oligopoly competition model with horizontal product differentiation. We analyze the equilibrium prices and how prices affect firm profits. We find product differentiation and the level of product substitutability affect prices. We also find that content price plays a significant role in affecting the price of the digital device. Content price can either increase or decrease the tied digital device price depending on the profit margin and demand elasticity of the digital content. We further analyze how content and device prices affect their respective profits and the overall profit of the firm. We extend our model to vertical product differentiation and find vertical product differentiation and the level of product quality affect prices.     

Optimal software pricing in the presence of piracy and word-of-mouth effect

We develop an analytical model that embeds empirical findings on software diffusion to examine optimal pricing strategies for a spreadsheet software product under coalescing effects of piracy and word-of-mouth through its entire life cycle. We find that the demand of the innovators has the most significant impact on the firm's pricing decision. Our research recommends market skimming pricing strategy if innovators' demand is high and the market penetration pricing strategy is preferred otherwise. Surprisingly, the increase of conversion rate of imitators to buyers never significantly alters the pricing strategy pre-determined by the demand of innovators. Most interestingly, the optimal profit from instituting a two prices policy for a software product with five years lifespan outperforms that from a one price policy by no more than 4%, a finding that corroborates the common one price policy observed in reality.     

An empirical evaluation of the comprehensibility of decision table, tree and rule based predictive models

An important objective of data mining is the development of predictive models. Based on a number of observations, a model is constructed that allows the analysts to provide classifications or predictions for new observations. Currently, most research focuses on improving the accuracy or precision of these models and comparatively little research has been undertaken to increase their comprehensibility to the analyst or end-user. This is mainly due to the subjective nature of ‘comprehensibility’, which depends on many factors outside the model, such as the user's experience and his/her prior knowledge. Despite this influence of the observer, some representation formats are generally considered to be more easily interpretable than others. In this paper, an empirical study is presented which investigates the suitability of a number of alternative representation formats for classification when interpretability is a key requirement. The formats under consideration are decision tables, (binary) decision trees, propositional rules, and oblique rules. An end-user experiment was designed to test the accuracy, response time, and answer confidence for a set of problem-solving tasks involving the former representations. Analysis of the results reveals that decision tables perform significantly better on all three criteria, while post-test voting also reveals a clear preference of users for decision tables in terms of ease of use.     

An evolutionary real options framework for the design and management of projects and systems with complex real options and exercising conditions

To address the issue of decision support for designing and managing flexible projects and systems in the face of uncertainties, this paper integrates real options valuation, decision analysis techniques, Monte Carlo simulations and evolutionary algorithms in an evolutionary real options framework. The proposed evolutionary real options framework searches for an optimized portfolio of real options and makes adaptive plans to cope with uncertainties as the future unfolds. Exemplified through a test case, the evolutionary framework not only compares favorably with traditional fixed design approaches but also delivers considerable improvements over prevailing real options practices.     

A framework for dynamic multiple-criteria decision making

The classic multiple-criteria decision making (MCDM) model assumes that, when taking a decision, the decision maker has defined a fixed set of criteria and is presented with a clear picture of all available alternatives. The task then reduces to computing the score of each alternative, thus producing a ranking, and choosing the one that maximizes this value.However, most real-world decisions take place in a dynamic environment, where the final decision is only taken at the end of some exploratory process. Exploration of the problem is often beneficial, in that it may unveil previously unconsidered alternatives or criteria, as well as render some of them unnecessary.In this paper we introduce a flexible framework for dynamic MCDM, based on the classic model, that can be applied to any dynamic decision process and which is illustrated by means of a small helicopter landing example. In addition, we outline a number of possible applications in very diverse fields, to highlight its versatility.     

Combination of sources of evidence with different discounting factors based on a new dissimilarity measure

The sources of evidence may have different reliability and importance in real applications for decision making. The estimation of the discounting (weighting) factors when the prior knowledge is unknown have been regularly studied until recently. In the past, the determination of the weighting factors focused only on reliability discounting rule and it was mainly dependent on the dissimilarity measure between basic belief assignments (bba's) represented by an evidential distance. Nevertheless, it is very difficult to characterize efficiently the dissimilarity only through an evidential distance. Thus, both a distance and a conflict coefficient based on probabilistic transformations BetP are proposed to characterize the dissimilarity. The distance represents the difference between bba's, whereas the conflict coefficient reveals the divergence degree of the hypotheses that two belief functions strongly support. These two aspects of dissimilarity are complementary in a certain sense, and their fusion is used as the dissimilarity measure. Then, a new estimation method of weighting factors is presented by using the proposed dissimilarity measure. In the evaluation of weight of a source, both its dissimilarity with other sources and their weighting factors are considered. The weighting factors can be applied in the both importance and reliability discounting rules, but the selection of the adapted discounting rule should depend on the actual application. Simple numerical examples are given to illustrate the interest of the proposed approach.     

Quickly locating efficient, equitable deals in automated negotiations under two-sided information uncertainty

This paper develops an automated negotiation procedure inclusive of mechanism design and agent design for bilateral multi-issue negotiations under two-sided information uncertainty. The proposed negotiation mechanism comprises a protocol called MUP (Monotonic Utility-granting Protocol) and a matching strategy called WYDIWYG (What You Display Influences What You Get). The proposed preference elicitation procedure makes the agents faithful surrogates of the user they represent while the proposed Frontier Tracking Proposal Construction Algorithm (FTPCA) makes them learn the opponent's flexibility in negotiation and respond appropriately. The mechanism design and the agent design together help in locating efficient and equitable deals quickly. The efficiency, stability, simplicity, distribution symmetry and incentive compatibility of the proposed procedure are demonstrated through negotiation simulation experiments.     

Cloud computing — The business perspective

The evolution of cloud computing over the past few years is potentially one of the major advances in the history of computing. However, if cloud computing is to achieve its potential, there needs to be a clear understanding of the various issues involved, both from the perspectives of the providers and the consumers of the technology. While a lot of research is currently taking place in the technology itself, there is an equally urgent need for understanding the business-related issues surrounding cloud computing. In this article, we identify the strengths, weaknesses, opportunities and threats for the cloud computing industry. We then identify the various issues that will affect the different stakeholders of cloud computing. We also issue a set of recommendations for the practitioners who will provide and manage this technology. For IS researchers, we outline the different areas of research that need attention so that we are in a position to advice the industry in the years to come. Finally, we outline some of the key issues facing governmental agencies who, due to the unique nature of the technology, will have to become intimately involved in the regulation of cloud computing.