Volumen 6 (2016): Heft 1 (January 2016)

In recent years, artificial neural networks have been commonly used for time series forecasting by researchers from various fields. There are some types of artificial neural networks and feed forward artificial neural networks model is one of them. Although feed forward artificial neural networks gives successful forecasting results they have a basic problem. This problem is architecture selection problem. In order to eliminate this problem, Yadav et al. (2007) proposed multiplicative neuron model artificial neural network. In this study, differential evolution algorithm is proposed for the training of multiplicative neuron model for forecasting. The proposed method is applied to two well-known different real world time series data.

¹This paper first introduces a new approach for studying bi-capacities and the bipolar Choquet integrals based on ternary-element sets. In the second half of the paper, we extend our approach to bi-capacities on fuzzy sets. Then, we propose a model of bipolar Choquet integral with respect to bi-capacities on fuzzy sets, and we give some basic properties of this model.

Traditional clustering algorithms are no longer suitable for use in data mining applications that make use of large-scale data. There have been many large-scale data clustering algorithms proposed in recent years, but most of them do not achieve clustering with high quality. Despite that Affinity Propagation (AP) is effective and accurate in normal data clustering, but it is not effective for large-scale data. This paper proposes two methods for large-scale data clustering that depend on a modified version of AP algorithm. The proposed methods are set to ensure both low time complexity and good accuracy of the clustering method. Firstly, a data set is divided into several subsets using one of two methods random fragmentation or K-means. Secondly, subsets are clustered into K clusters using K-Affinity Propagation (KAP) algorithm to select local cluster exemplars in each subset. Thirdly, the inverse weighted clustering algorithm is performed on all local cluster exemplars to select well-suited global exemplars of the whole data set. Finally, all the data points are clustered by the similarity between all global exemplars and each data point. Results show that the proposed clustering method can significantly reduce the clustering time and produce better clustering result in a way that is more effective and accurate than AP, KAP, and HAP algorithms.

The purpose of this study was to generate more concise rule extraction from the Recursive-Rule Extraction (Re-RX) algorithm by replacing the C4.5 program currently employed in Re-RX with the J48graft algorithm. Experiments were subsequently conducted to determine rules for six different two-class mixed datasets having discrete and continuous attributes and to compare the resulting accuracy, comprehensibility and conciseness. When working with the CARD1, CARD2, CARD3, German, Bene1 and Bene2 datasets, Re-RX with J48graft provided more concise rules than the original Re-RX algorithm. The use of Re-RX with J48graft resulted in 43.2%, 37% and 21% reductions in rules in the case of the German, Bene1 and Bene2 datasets compared to Re-RX. Furthermore, the Re-RX with J48graft showed 8.87% better accuracy than the Re-RX algorithm for the German dataset. These results confirm that the application of Re-RX in conjunction with J48graft has the capacity to facilitate migration from existing data systems toward new concise analytic systems and Big Data.

Portfolio optimization, one of the most rapidly growing field of modern finance, is selection process, by which investor chooses the proportion of different securities and other assets to held. This paper studies the influence of membership function’s shape on the result of fuzzy portfolio optimization and focused on portfolio selection problem based on credibility measure. Four different shapes of the membership function are examined in the context of the most popular optimization problems: mean-variance, mean-semivariance, entropy minimization, value-at-risk minimization. The analysis takes into account both: the study of necessary and sufficient conditions for the existence of extremes, as well as the statistical inference about the differences based on simulation.