Search

To link to full-text access for this article, visit this link: http://dx.doi.org/10.1016/j.ejor.2006.06.031 Byline: Young U. Ryu (a), R. Chandrasekaran (b), Varghese S. Jacob (c) Keywords: Data mining; Isotonic separation; Breast cancer diagnosis Abstract: 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. Author Affiliation: (a) School of Management, The University of Texas at Dallas, P.O. Box 830688, SM 33, Richardson, TX 75083-0688, USA (b) School of Engineering and Computer Science, The University of Texas at Dallas, P.O. Box 830688, Richardson, TX 75083-0688, USA (c) School of Management, The University of Texas at Dallas, P.O. Box 830688, SM 40, Richardson, TX 75083-0688, USA Article History: Received 18 February 2005; Accepted 5 June 2006

For the task-scheduling problem, we propose an augmented neural-network approach, which allows the integration of greedy as well as nongreedy heuristics (AugNN-GNG), to give improved solutions in a small number […]

To link to full-text access for this article, visit this link: http://dx.doi.org/10.1016/j.ejor.2005.03.045 Byline: Anurag Agarwal (a), Selcuk Colak (a), Varghese S. Jacob (b), Hasan Pirkul (b) Keywords: Scheduling; Neural networks; Heuristics Abstract: We propose new heuristics along with an augmented-neural-network (AugNN) formulation for solving the makespan minimization task-scheduling problem for the non-identical machine environment. We explore four task and three machine-priority rules, resulting in 12 combinations of single-pass heuristics. The task-priority rules are Highest-Level-First (HLF), Highest-Total-Remaining-Processing-Time-First (HTRPTF), Smallest-Latest-Finish-Time-First (SLFTF) and Minimum-Slack-First (MSF). For machine priority, we propose a greedy rule called Fastest-Available-Machine-First (FAMF), and two non-greedy rules: (1) Fastest-Available-Machine-First-With-Conditional-Wait-1 (FAMF-CW-1), and (2) Fastest-Available-Machine-First-With-Conditional-Wait-2 (FMF-CW-2). The AugNN approach integrates neural-network learning with domain and problem specific knowledge through heuristics, to produce good results. A single-pass heuristic is embedded in a neural network designed specifically for each problem. We give the AugNN formulation for each of the 12 heuristics and show computational results on 100 randomly generated problems of sizes ranging from 20 to 70 tasks and 2 to 5 machines. Results demonstrate that AugNN provides significant improvement over single-pass heuristics. The reduction in the gap between the obtained solution and the lower-bound due to AugNN over single-pass heuristics ranged from 24.4% to 50%. As far as heuristics, the non-greedy machine-priority rules performed significantly better than the greedy rule. The average gaps for the non-greedy rules ranged from 16.1% to 23.5% compared to 33.7% to 40.4% for greedy. For AugNN, for non-greedy rules, the gap ranged from 11.5% to 15.5% compared to 18.4% to 25.0% for greedy. The HLF and HTRPTF task priority rules performed better than the other two. The HTRPTF/FAMF-CW-1 combination gave the best results, closely followed by HLF/FAMF-CW-2 and HTRPTF/FAMF-CW-2 combinations. Author Affiliation: (a) Department of Decision and Information Sciences, Warrington College of Business, University of Florida, Gainesville, FL 32611-7169, USA (b) School of Management, University of Texas at Dallas, Richardson TX 75083-0688, USA Article History: Received 6 June 2003; Accepted 14 March 2005

To link to full-text access for this article, visit this link: http://dx.doi.org/10.1016/j.ejor.2005.07.005 Byline: Subodha Kumar (a), Varghese S. Jacob (b), Chelliah Sriskandarajah (b) Keywords: Web advertising; Scheduling; NP-hard; Heuristics; Genetic algorithms; Bin-packing Abstract: Many web sites (e.g. Hotmail, Yahoo) provide free services to the users while generating revenues from advertising. Advertising revenue is, therefore, critical for these sites. This in turn raises the question, how should advertisements at a web site be scheduled in a planning horizon to maximize revenue. Advertisements on the web are specified by geometry and display frequency and both of these factors need to be considered in developing a solution to the advertisement scheduling problem. Since this problem belongs to the class of NP-hard problems, we first develop a heuristic called LSMF to solve the problem. This heuristic is then combined with a genetic algorithm (GA) to develop a hybrid GA. The hybrid GA solution is first compared with the upper bound obtained by running CPLEX for the integer programming formulation of the problem. It is then also compared with an existing algorithm proposed in the literature. Our computational results show that the hybrid GA performs exceptionally well in the sense that it provides optimal or near optimal solution for a wide range of problem instances of realistic sizes and the improvements over existing algorithm are substantial. Finally we present a case study to illustrate how revenue could be significantly increased with a small improvement in the advertisement schedule. It is the first such study in this setup. Author Affiliation: (a) Department of Management Science, University of Washington Business School, Box 353200, Seattle, WA 98195, USA (b) Management Science and Information Systems, School of Management, University of Texas at Dallas, Box 830688, Mail Station JO4.7, Richardson, TX 75083-0688, USA

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 […]

In this paper, we investigate the efficacy of artificial intelligence (AI) based meta-heuristic techniques namely genetic algorithms (GAs), for the product line design problem. This work extends previously developed methods for the single product design problem. We conduct a large scale simulation study to determine the effectiveness of such an AI based technique for providing good solutions and bench mark the performance of this against the current dominant approach of beam search (BS). We investigate the potential advantages of pursuing the avenue of developing hybrid models and then implement and study such hybrid models using two very distinct approaches: namely, seeding the initial GA population with the BS solution, and employing the BS solution as part of the GA operator's process. We go on to examine the impact of two alternate string representation formats on the quality of the solutions obtained by the above proposed techniques. We also explicitly investigate a critical managerial factor of attribute importance in terms of its impact on the solutions obtained by the alternate modeling procedures. The alternate techniques are then evaluated, using statistical analysis of variance, on a fairy large number of data sets, as to the quality of the solutions obtained with respect to the state-of-the-art benchmark and in terms of their ability to provide multiple, unique product line options. Index Terms--AI, attribute importance, beam search, GA, hybrid genetic algorithms, marketing, meta-heuristic techniques.

A new approach, called Augmented Neural Networks (AugNN) is proposed for solving the task-scheduling problem. The approach is a hybrid of the heuristic and the neural networks approaches.

The Frequency-Sensitive Competitive Learning (FSCL) neural network and the K-means non-hierarchical clustering method were applied to a cluster analysis of coffee brand choice data to evaluate their performances. Specifically, simulated data and actual brand choice probabilities were compared. The results showed that the combination of the FSCL network and K-means clustering produced the best results in terms of face validity, disparity in cluster sizes and ease of interpretation.

Optimal product design is acknowledged as a critical undertaking that can greatly affect the performance and competitiveness of firms. The use of an adaptive search technique known as Genetic Algorithms (GA) is proposed in solving product design problems. The performance of this approach is evaluated and compared to that of the dynamic programming heuristic of Kohli and Krishnamurti (1987) to product designs with attributes having different numbers of levels and alternative attribute sequencing rules. Results across 192 data sets reveal that the GA technique generates solutions which are 99.13% and 99.92% of the optimal product profile when the goals of the problem are to maximize the share-of-choices and the buyers' welfare, respectively. The figures are statistically better than the solutions identified by the dynamic programming heuristic of Kohli and Krishnamurti (1987, 1989) by 3% and 1% for the two objectives.

Existing computerized systems that support scheduling decisions for flexible manufacturing systems (FMS's) rely largely on knowledge acquired through rote learning (i.e., memorization) for schedule generation. In a few instances, the systems also possess some ability to learn using deduction or supervised induction. We introduce a novel AI-based system for generating static schedules that makes heavy use of an unsupervised learning module in acquiring significant portions of the requisite problem processing knowledge. This scheduler pursues a hybrid schedule generation strategy wherein it effectively combines knowledge acquired via genetics-based unsupervised induction with rote-learned knowledge in generating high-quality schedules in an efficient manner. Through a series of experiments conducted on a randomly generated problem of practical complexity, we show that the hybrid scheduler strategy is viable, promising, and, worthy of more in-depth investigations.

Cloud computing can be viewed as an e‐business in which the cloud provider sells a range of information technology services and operations over the Internet to firms. Security is a major concern for firms that move to the cloud. The security of a cloud is the joint responsibility of the cloud provider and cloud users. However, the extent to which the provider and users can affect cloud security through their efforts depends on the cloud service model. In this study, we develop a game‐theoretical model to study the impact of the cloud service model on provider and user incentives to exert security effort. Our results show that for a given service model, an increase in the user loss from a security breach induces users to exert more security effort. However, if the provider's service cost is low, the provider profitably free rides on the enhanced user incentives to exert security effort by diminishing his effort. Analogously, the cloud provider is able to profitably free ride on users' enhanced incentives to exert security effort when the user population is more homogeneous in terms of either cloud valuation or loss from a security breach, depending on the service cost. Our results also have implications for how a cloud provider could possibly target a particular cloud service model to specific user groups based on their characteristics and likelihood of security loss. [ABSTRACT FROM AUTHOR]

Pay-per-click (PPC) is a common pricing model used to pay for ads on the Web and is open to the possibility for click fraud, where clicks are not from a legitimate user. Identifying click fraud is generally done in a three-stage process: the service provider (SP) first classifies clicks as fraudulent or not, then the advertiser does the same with a different technology, and if there is a disagreement, the SP examines further and his conclusions are considered binding. The advertiser pays for clicks that are identified as valid in the first two stages or confirmed as valid in the last stage. We model the choice of the identification technologies as a double moral hazard problem. We analyze the case where the PPC is incentive compatible to overcome the moral hazard problem, and examine the question of whether the incentive compatible PPC is sufficient to incentivize the two parties to unilaterally make further improvements to their identification technologies and simultaneously increase their profits. We show that when the cost of the third-stage identification technology is large, which is likely to be the case because of its complexity and use of expensive human experts, the incentive compatible PPC does not support unilateral technological improvements. We then examine a setting where the third-stage identification is delegated to a third party and find that this arrangement can induce unilateral improvements to the identification technologies in the first two stages. Collectively our results show that although the PPC model itself may not induce improvements in the first two stages of click fraud identification, a common arrangement espoused of having a third party resolve disagreements helps make PPC support unilateral technological improvements. Accordingly, we show an indirect benefit to the third-party arrangement. [ABSTRACT FROM AUTHOR]