Ekrem Duman

Migrating Birds Optimization: A new metaheuristic approach and its performance on quadratic assignment problem

We propose a new nature inspired metaheuristic approach based on the V flight formation of the migrating birds which is proven to be an effective formation in energy saving. Its performance is tested on quadratic assignment problem instances arising from a real life problem and very good results are obtained. The quality of the solutions we report are better than simulated annealing, tabu search, genetic algorithm, scatter search, particle swarm optimization, differential evolution and guided evolutionary simulated annealing approaches. The proposed method is also tested on a number of benchmark problems obtained from the QAPLIB and in most cases it was able to obtain the best known solutions. These results indicate that our new metaheuristic approach could be an important player in metaheuristic based optimization.

Detecting credit card fraud by genetic algorithm and scatter search

In this study we develop a method which improves a credit card fraud detection solution currently being used in a bank. With this solution each transaction is scored and based on these scores the transactions are classified as fraudulent or legitimate. In fraud detection solutions the typical objective is to minimize the wrongly classified number of transactions. However, in reality, wrong classification of each transaction do not have the same effect in that if a card is in the hand of fraudsters its whole available limit is used up. Thus, the misclassification cost should be taken as the available limit of the card. This is what we aim at minimizing in this study. As for the solution method, we suggest a novel combination of the two well known meta-heuristic approaches, namely the genetic algorithms and the scatter search. The method is applied to real data and very successful results are obtained compared to current practice.

A cost-sensitive decision tree approach for fraud detection

With the developments in the information technology, fraud is spreading all over the world, resulting in huge financial losses. Though fraud prevention mechanisms such as CHIP&PIN are developed for credit card systems, these mechanisms do not prevent the most common fraud types such as fraudulent credit card usages over virtual POS (Point Of Sale) terminals or mail orders so called online credit card fraud. As a result, fraud detection becomes the essential tool and probably the best way to stop such fraud types. In this study, a new cost-sensitive decision tree approach which minimizes the sum of misclassification costs while selecting the splitting attribute at each non-terminal node is developed and the performance of this approach is compared with the well-known traditional classification models on a real world credit card data set. In this approach, misclassification costs are taken as varying. The results show that this cost-sensitive decision tree algorithm outperforms the existing well-known methods on the given problem set with respect to the well-known performance metrics such as accuracy and true positive rate, but also a newly defined cost-sensitive metric specific to credit card fraud detection domain. Accordingly, financial losses due to fraudulent transactions can be decreased more by the implementation of this approach in fraud detection systems.

The quadratic assignment problem in the context of the printed circuit board assembly process

The sequencing of placement and the configuration of the feeder is among the main problems involved in printed circuit board (PCB) assembly optimization. In some machine architectures, the latter problem can be formulated as the well known NP-hard quadratic assignment problem (QAP). In this study, a search is made among those metaheuristics that have recently found widespread application in order to identify a heuristic procedure that performs well with the QAP in the PCB assembly context. To this end, specific algorithms reflecting implementations of Taboo Search, Simulated Annealing and Genetic Algorithm-type metaheuristics were tested and compared using real PCB assembly data. The same set of algorithms was also tested on general QAP problems and it was observed that algorithms which performed successfully in the PCB context could perform poorly in the general situation. In the light of this, it can be concluded that ascertaining the best performing heuristic is complicated by the fact that the performance of a heuristic depends on the context of the problem, which determines the structure and relationships of problem parameters.

Precedence constrained TSP arising in printed circuit board assembly

Component placement sequencing is a challenging problem that arises in automated assembly of printed circuit boards. While for some placement machines all placement sequences are acceptable, in other cases some sequences are not allowed because of the shape of the placement head. In such cases, while the head moves down to perform a placement, it might damage a previously placed component, and the problem of determining a minimum cost and at the same time acceptable sequence leads to a Precedence Constrained Travelling Salesman Problem formulation. In this study, a solution procedure to such a formulation is developed and its implementation in a real PCB assembly environment is discussed.

Detecting credit card fraud by Modified Fisher Discriminant Analysis

We introduce Fisher Linear Discriminant Analysis (FLDA).We modify it to be sensitive toward profitable instances.We applied them together in credit card fraud detection problem.The results are compared in terms of total obtained profit with three well-known models.Modified fisher outperforms the other models in attaining high profit. In parallel to the increase in the number of credit card transactions, the financial losses due to fraud have also increased. Thus, the popularity of credit card fraud detection has been increased both for academicians and banks. Many supervised learning methods were introduced in credit card fraud literature some of which bears quite complex algorithms. As compared to complex algorithms which somehow over-fit the dataset they are built on, one can expect simpler algorithms may show a more robust performance on a range of datasets. Although, linear discriminant functions are less complex classifiers and can work on high-dimensional problems like credit card fraud detection, they did not receive considerable attention so far. This study investigates a linear discriminant, called Fisher Discriminant Function for the first time in credit card fraud detection problem. On the other hand, in this and some other domains, cost of false negatives is very higher than false positives and is different for each transaction. Thus, it is necessary to develop classification methods which are biased toward the most important instances. To cope for this, a Modified Fisher Discriminant Function is proposed in this study which makes the traditional function more sensitive to the important instances. This way, the profit that can be obtained from a fraud/legitimate classifier is maximized. Experimental results confirm that Modified Fisher Discriminant could eventuate more profit.

Detecting credit card fraud by ANN and logistic regression

With the developments in information technology and improvements in communication channels, fraud is spreading all over the world, resulting in huge financial losses. Though fraud prevention mechanisms such as CHIP&PIN are developed, these mechanisms do not prevent the most common fraud types such as fraudulent credit card usages over virtual POS terminals through Internet or mail orders. As a result, fraud detection is the essential tool and probably the best way to stop such fraud types. In this study, classification models based on Artificial Neural Networks (ANN) and Logistic Regression (LR) are developed and applied on credit card fraud detection problem. This study is one of the firsts to compare the performance of ANN and LR methods in credit card fraud detection with a real data set.

Comparing alternative classifiers for database marketing: The case of imbalanced datasets

There are various algorithms used for binary classification where the cases are classified into one of two non-overlapping classes. The area under the receiver operating characteristic (ROC) curve is the most widely used metric to evaluate the performance of alternative binary classifiers. In this study, for the application domains where the high degree of imbalance is the main characteristic and the identification of the minority class is more important, we show that hit rate based measures are more correct to assess model performances and that they should be measured on out of time samples. We also try to identify the optimum composition of the training set. Logistic regression, neural network and CHAID algorithms are implemented for a real marketing problem of a bank and the performances are compared.

Dynamic simulation for strategic insurance management

In this paper, a dynamic model-based management consultancy project carried out for a major insurance company in Turkey is presented. The objective of the project was to address certain strategic managerial problems of the company by using systemic dynamic simulation. The main strategic problem of concern was that the company exhibited a fast growth between 1988 and 1993, followed by persistent stagnation and even a slight decline. This paper describes the main structures of the model, presents the validity tests and lists the major results of the study. The model is developed, calibrated and validated using real data for seven years, between 1989 and 1996. The main benefit of the model is that it generates a systemic and dynamic understanding of the company’s internal and external interactions so as to enable creative solutions for existing and potential problems. One of the recommendations of the project has actually been initiated as a pilot project. A new interactive gaming version of the model is in the final stages of completion. The model and the game version can be used as a ‘‘learning laboratory’’ in the company, which would be a first step toward ‘‘organizational learning’’. Copyright

Optimization issues in automated production of printed circuit boards: operations sequencing, feeder configuration and load balancing problems

In electronics industry, the widely used automatic placement machines insert or amount electronic components, supplied from sequential or random access feeders, to predefined locations on printed circuit boards. The sequencing of placement operations, the assignment of component types to feeder cells and load balancing in these machines directly influence productivity. In this research, such sequencing, assignment and balancing problems, arising under various machine architectures, are modeled based on a decomposition leading to travelling salesman, rural postman, assignment, quadratic assignment and/or line balancing problems. For each machine architecture a specific heuristic solution procedure, involving iterative deployment and solution of two or more of these problems is suggested. Results are presented for some test problems. An implementation in a production facility is also discussed.