Young U. Ryu

Firm bankruptcy prediction: experimental comparison of isotonic separation and other classification approaches

A newly introduced method called isotonic separation is evaluated in the prediction of firm bankruptcy. Feature reduction methods are first applied to reduce the ratios used in the prediction. Then, various classification methods, including discriminant analysis, neural networks, decision tree induction, learning vector quantization, rough sets, and isotonic separation, are used with the reduced ratios. Experiments show that the isotonic separation method is a viable technique, performing generally better than other methods for short-term bankruptcy prediction.

Network externalities, layered protection and IT security risk management

This paper considers two important issues related to security risk management. First, the presence of network externalities in security risks. Second, the distinction of general (network) and system-specific protection measures. We found the optimal allocation of security resources (investments) in protecting every system in an organization. The results show that the consideration of network externalities and layered protection changes the risk mitigation decisions significantly. In addition, accurate estimation of system risk plays a critical role in the success of risk management. Otherwise, the use of a uniform baseline protection approach may be more desirable when the misjudgment of relative system risks is likely to occur.

Breast cancer prediction using the isotonic separation technique

A recently developed data separation/classification method, called isotonic separation, is applied to breast cancer prediction. Two breast cancer data sets, one with clean and sufficient data and the other with insufficient data, are used for the study and the results are compared against those of decision tree induction methods, linear programming discrimination methods, learning vector quantization, support vector machines, adaptive boosting, and other methods. The experiment results show that isotonic separation is a viable and useful tool for data classification in the medical domain. 2006 Elsevier B.V. All rights reserved.

Unrealistic optimism on information security management

Information security is a critical issue that many firms face these days. While increasing incidents of information security breaches have generated extensive publicity, previous studies repeatedly expose low levels of managerial awareness and commitment, a key obstacle to achieving a good information security posture. The main motivation of our study emanates from this phenomenon that the increased vulnerability to information security breaches is coupled with the low level of managerial awareness and commitment regarding information security threats. We report this dissonance by addressing a cognitive bias called optimistic bias. Using a survey, we study if MIS executives are subject to such a bias in their vulnerability perceptions of information security. We find that they demonstrate optimistic bias in risk perception on information security domain. The extent of this optimistic bias is greater with a distant comparison target with fewer information sharing activities. This optimistic bias is also found to be related to perception of controllability with information security threats. In order to overcome the effects of optimistic bias, firms need more security awareness training and systematic treatments of security threats instead of relying on ad hoc approach to security measure implementation.

Internet content filtering using isotonic separation on content category ratings

The World Wide Web has enabled anybody with a low cost Internet connection to access vast information repositories. Some of these repositories contain information (e.g., hate speech and pornography) that is considered objectionable, especially for children to view. Several efforts---legal and technical---are underway to protect children and the generic public from accessing this type of content. We propose a technical approach utilizing a recently proposed technique called isotonic separation for filtering with content metadata if they satisfy monotone conditions. We illustrate this approach using a category rating method of PICS. In essence, we formulate the Internet content filtering problem as a classification problem on content metadata and report on experiments we conducted with the isotonic separation technique.

Isotonic Separation

Data classification and prediction problems are prevalent in many domains. The need to predict to which class a particular data point belongs has been seen in areas such as medical diagnosis, credit rating, Web filtering, prediction, and stock rating. This has led to strong interest in developing systems that can accurately classify data and predict outcome. The classification is typically based on the feature values of objects being classified. Often, a form of ordering relation, defined by feature values, on the objects to be classified is known. For instance, the objects belonging to one class have larger (or smaller) feature values than do those in the other class. Exploiting this characteristic of isotonicity, we propose a data-classification method called isotonic separation based on linear programming, especially network programming. The paper also addresses an extension of the isotonic-separation method for continuous outcome prediction. Applications of the isotonic separation for discrete outcome prediction and its extension for continuous outcome prediction are shown to illustrate its applicability.

Exploration, exploitation and adaptive rationality: the neo-Schumpeterian perspective

Abstract Resource allocation between exploration of emerging technological possibilities and exploitation of known technological possibilities involves a delicate trade-off. We develop a model to represent this trade-off under the time-pressing situation where the firm’s existing basis of survival is constantly challenged by competitors’ innovation and imitation. We examine how the employment of an adaptive rule improves a balance between the exploration and the exploitation. Simulation experiments show that an adaptively rational decision rule, or a step-by-step exploration of unknown opportunities based on feedback on returns, is more likely to increase firm survival under diverse conditions than an all-or-nothing approach regarding the unknown opportunities. Furthermore, our study suggests that the adaptively rational rule is self-protected from too much loss, while its potential pay-off can be unbounded above.

A hierarchical constraint satisfaction approach to product selection for electronic shopping support

The development of advanced information and telecommunications technologies has established more convenient ways of interorganizational business transactions. Especially, various forms of electronic trading systems are introduced, which replicate, and often improve, functions of physical market places. We propose a product selection mechanism for such an electronic marketplace, which is viewed as a satisfaction problem of hierarchically organized constraints over product attributes. The proposed approach is more expressively powerful and flexible than product selection based on a single product taxonomy hierarchy.

Classification algorithm sensitivity to training data with non representative attribute noise

We present an empirical comparison of classification algorithms when training data contains attribute noise levels not representative of field data. To study algorithm sensitivity, we develop an innovative experimental design using noise situation, algorithm, noise level, and training set size as factors. Our results contradict conventional wisdom indicating that investments to achieve representative noise levels may not be worthwhile. In general, over representative training noise should be avoided while under representative training noise is less of a concern. However, interactions among algorithm, noise level, and training set size indicate that these general results may not apply to particular practice situations.

IPM-EPDL: an XML-based executable process definition language

The capability to quickly design and deploy a new business process has become critical for companies to succeed in the rapidly changing global market. Web services are considered to be the most promising technology that enables automatic execution of business processes. It has some limitations, however, such as lack support for organizational structure and role model, customized user interface, access authorization of data, and storing instance data. This paper introduces Integrated Process Management Executable Process Definition Language (IPM-EPDL), an XML-based executable process definition language. IPM-EPDL supports flexible specification of organizational structure and role model. It allows the user to design the interface of a business process activity according to the number of users as well as their preferences, browser type, and type of the form to be used for the activity. It also enables to specify the access authorization of data and to check the validity of the data. Finally, it can specify all business data and keep it throughout the entire process lifecycle. Such data can be used for checking the status of process instance and for various analyses. The paper discusses how a typical process definition must be extended to enable its execution while providing the above advantages. The paper describes how the IPM-EPDL execution engine provides these capabilities. The paper also demonstrates a prototype system along with an extended example to illustrate that the proposed language and process execution engine can support rapid deployment of a new business process that can be performed on various platforms even by a remote user.