The aim of our research was to apply well-known data mining techniques (such as linear neural networks, multi-layered perceptrons, probabilistic neural networks, classification and regression trees, support vector machines and finally a hybrid decision tree neural network approach) to the problem of predicting the quality of service in call centers; based on the performance data actually collected in a call center of a large insurance company. Our aim was two-fold. First, to compare the performance of models built using the above-mentioned techniques and, second, to analyze the characteristics of the input sensitivity in order to better understand the relationship between the perform-ance evaluation process and the actual performance and in this way help improve the performance of call centers. In this paper we summarize our findings.
Past few years have witnessed a growing recognition of intelligent techniques for the construction of efficient and reliable intrusion detection systems. Due to increasing incidents of cyber attacks, building effective intrusion detection systems (IDS) are essential for protecting information systems security, and yet it remains an elusive goal and a great challenge. In this paper, we report a performance analysis between Multivariate Adaptive Regression Splines (MARS), neural networks and support vector machines. The MARS procedure builds flexible regression models by fitting separate splines to distinct intervals of the predictor variables. A brief comparison of different neural network learning algorithms is also given.
Sorting by reversals is an important problem in inferring the evolutionary relationship between two genomes. The problem of sorting unsigned permutation has been proven to be NP-hard. The best guaranteed error bounded is the 3/2- approximation algorithm. However, the problem of sorting signed permutation can be solved easily. Fast algorithms have been developed both for finding the sorting sequence and finding the reversal distance of signed permutation. In this paper, we present a way to view the problem of sorting unsigned permutation as signed permutation. And the problem can then be seen as searching an optimal signed permutation in all n2 corresponding signed permutations. We use genetic algorithm to conduct the search. Our experimental result shows that the proposed method outperform the 3/2-approximation algorithm.
Classification of texture pattern is one of the most important problems in pattern recognition. In this paper, we present a classification method based on the Discrete Cosine Transform (DCT) coefficients of texture image. As DCT works on gray level image, the color scheme of each image is transformed into gray levels. For classifying the images using DCT we used two popular soft computing techniques namely neurocomputing and neuro-fuzzy computing. We used a feedforward neural network trained using the backpropagation learning and an evolving fuzzy neural network to classify the textures. The soft computing models were trained using 80% of the texture data and remaining was used for testing and validation purposes. A performance comparison was made among the soft computing models for the texture classification problem. We also analyzed the effects of prolonged training of neural networks. It is observed that the proposed neuro-fuzzy model performed better than neural network.
Long-term rainfall prediction is a challenging task especially in the modern world where we are facing the major environmental problem of global warming. In general, climate and rainfall are highly non-linear phenomena in nature exhibiting what is known as the butterfly effect. While some regions of the world are noticing a systematic decrease in annual rainfall, others notice increases in flooding and severe storms. The global nature of this phenomenon is very complicated and requires sophisticated computer modeling and simulation to predict accurately. In this paper, we report a performance analysis for Multivariate Adaptive Regression Splines (MARS)and artificial neural networks for one month ahead prediction of rainfall. To evaluate the prediction efficiency, we made use of 87 years of rainfall data in Kerala state, the southern part of the Indian peninsula situated at latitude -longitude pairs (8o29'N - 76o57' E). We used an artificial neural network trained using the scaled conjugate gradient algorithm. The neural network and MARS were trained with 40 years of rainfall data. For performance evaluation, network predicted outputs were compared with the actual rainfall data. Simulation results reveal that MARS is a good forecasting tool and performed better than the considered neural network.
Fusion of Artificial Neural Networks (ANN) and Fuzzy Inference Systems (FIS) have attracted the growing interest of researchers in various scientific and engineering areas due to the growing need of adaptive intelligent systems to solve the real world problems. ANN learns from scratch by adjusting the interconnections between layers. FIS is a popular computing framework based on the concept of fuzzy set theory, fuzzy if-then rules, and fuzzy reasoning. The advantages of a combination of ANN and FIS are obvious. There are several approaches to integrate ANN and FIS and very often it depends on the application. We broadly classify the integration of ANN and FIS into three categories namely concurrent model, cooperative model and fully fused model. This paper starts with a discussion of the features of each model and generalize the advantages and deficiencies of each model. We further focus the review on the different types of fused neuro-fuzzy systems and citing the advantages and disadvantages of each model.
Neuro-fuzzy systems have attracted growing interest of researchers in various scientific and engineering areas due to the increasing need of intelligent systems. This paper evaluates the use of two popular soft computing techniques and conventional statistical approach based on Box--Jenkins autoregressive integrated moving average (ARIMA) model to predict electricity demand in the State of Victoria, Australia. The soft computing methods considered are an evolving fuzzy neural network (EFuNN) and an artificial neural network (ANN) trained using scaled conjugate gradient algorithm (CGA) and backpropagation (BP) algorithm. The forecast accuracy is compared with the forecasts used by Victorian Power Exchange (VPX) and the actual energy demand. To evaluate, we considered load demand patterns for 10 consecutive months taken every 30 min for training the different prediction models. Test results show that the neuro-fuzzy system performed better than neural networks, ARIMA model and the VPX forecasts.
The integration of different learning and adaptation techniques to overcome individual limitations and to achieve synergetic effects through the hybridization or fusion of these techniques has, in recent years, contributed to a large number of new intelligent system designs. Computational intelligence is an innovative framework for constructing intelligent hybrid architectures involving Neural Networks (NN), Fuzzy Inference Systems (FIS), Probabilistic Reasoning (PR) and derivative free optimization techniques such as Evolutionary Computation (EC). Most of these hybridization approaches, however, follow an ad hoc design methodology, justified by success in certain application domains. Due to the lack of a common framework it often remains difficult to compare the various hybrid systems conceptually and to evaluate their performance comparatively. This chapter introduces the different generic architectures for integrating intelligent systems. The designing aspects and perspectives of different hybrid archirectures like NN-FIS, EC-FIS, EC-NN, FIS-PR and NN-FIS-EC systems are presented. Some conclusions are also provided towards the end.
In this paper, we examine the performance of four fuzzy rule generation methods on Wisconsin breast cancer data. The first method generates fuzzy if then rules using the mean and the standard deviation of attribute values. The second approach generates fuzzy if then rules using the histogram of attributes values. The third procedure generates fuzzy if then rules with certainty of each attribute into homogeneous fuzzy sets. In the fourth approach, only overlapping areas are partitioned. The first two approaches generate a single fuzzy if then rule for each class by specifying the membership function of each antecedent fuzzy set using the information about attribute values of training patterns. The other two approaches are based on fuzzy grids with homogeneous fuzzy partitions of each attribute. The performance of each approach is evaluated on breast cancer data sets. Simulation results show that the Modified grid approach has a high classification rate of 99.73 %.
Self-organizing complex systems typically are comprised of a large number of frequently similar components or events. Through their process, a pattern at the global-level of a system emerges solely from numerous interactions among the lower-level components of the system. Moreover, the rules specifying interactions among the system's components are executed using only local information, without reference to the global pattern, which, as in many real-world problems is not easily accessible or possible to be found. Stigmergy, a kind of indirect communication and learning by the environment found in social insects is a well know example of self-organization, providing not only vital clues in order to understand how the components can interact to produce a complex pattern, as can pinpoint simple biological non-linear rules and methods to achieve improved artificial intelligent adaptive categorization systems, critical for Data-Mining. On the present work it is our intention to show that a new type of Data-Mining can be designed based on Stigmergic paradigms, taking profit of several natural features of this phenomenon. By hybridizing bio-inspired Swarm Intelligence with Evolutionary Computation we seek for an entire distributed, adaptive, collective and cooperative self-organized Data-Mining. As a real-world, real-time test bed for our proposal, World-Wide-Web Mining will be used. Having that purpose in mind, Web usage Data was collected from the Monash University's Web site (Australia), with over 7 million hits every week. Results are compared to other recent systems, showing that the system presented is by far promising.