nan Subanar

Clustering based-on indiscernibility and indiscernibility level

The core concept of classical rough sets are clustering similarities and dissimilarities of objects based on the notions of indiscernibility and discernibility. In this paper, we present a new method of clustering data based on the combination of indiscernibility and its indiscernibility level. The indiscernibility level quantifies the indiscernibility of pairs of objects among other objects in information systems. The result of this paper show the dual notions of indiscernibility and its indiscernibility level play an important role in clustering information systems.

Reducing Hierarchical Clustering Instability Using Clustering Based on Indiscernibility and Indiscernibility Level

The notions of indiscernibility and discernibility are the core concept of classical rough sets to cluster similarities and differences of data objects. In this paper, we use a new method of clustering data based on the combination of indiscernibility (quantitative indiscernibility relations) and its indiscernibility level. The indiscernibility level quantify the indiscernibility of pair of objects among other objects in information systems and this level represent the granularity of pair of objects in information system. For comparison to the new method, the following four clustering methods were selected and evaluated on a simulation data set : average-, complete- and single-linkage agglomerative hierarchical clustering and Ward’s method. The result of this paper shows that the four methods of hierarchical clustering yield dendrogram instability that give different solution under permutation of input order of data object while the new method reduce dendrogram instability.

WMEVF: An outlier detection methods for categorical data

Outliers are uncommon events in real life. For a database processing, an outlier means unusual record comparing to the others. An outlier can be caused by a damage to a system, an intruder in a system, or a new fact in a system. Outlier detection is an important task to find an exceptional data.

A Second Correlation Method for Multivariate Exchange Rates Forecasting

Foreign exchange market is one of the most complex dynamic market with high volatility, non linear and irregularity. As the globalization spread to the world, exchange rates forecasting become more important and complicated. Many external factors influence its volatility. To forecast the exchange rates, those external variables can be used and usually chosen based on the correlation to the predicted variable. A new second correlation method to improve forecasting accuracy is proposed. The second correlation is used to choose the external variable with different time interval. The proposed method is tested using six major monthly exchange rates with Nonlinear Autoregressive with eXogenous input (NARX) compared with Nonlinear Autoregressive (NAR) for model benchmarking. We evaluated the forecasting accuracy of proposed method is increasing by 16.8% compared to univariate NAR model and slight better than linear correlation on average for Dstat parameter and gives almost no improvement for MSE.

Multiresolution radial basis model for nonlinear time series prediction

The multiresolution radial basis autoregressive model for nonlinear time series prediction is proposed in this paper. This is a development form of the multiresolution autoregressive model. This constitutes an alternative method for nonlinear time series prediction, especially for threshold model. In the beginning, time series is decomposed by wavelet transform to produce smooth part and detail part. The both are representing the main pattern and noise pattern of time series, respectively. Some radial basis functions are performed to refine the decomposition result into homogenous clusters. This refining result becomes an input series for a neural network structure. The model is applied to make one step prediction ahead of a simulation data. The normality test for residual is accepted, and the statistic test shows that the model coefficients are significant.

Fractional integrated recurrent neural network (FIRNN) for forecasting of time series data in electricity load in Java-Bali

The Increasing demand for electricity causes problems for State Electricity Company (SEC), which is called PLN in Indonesia in providing services to the public. Neural network (NN) is one of the methods that is often used in forecasting electricity load in different countries. Another form of neural network which is widely used for the analysis of issues that have a repeating pattern is the model of Recurrent Neural Network (RNN). Particularly, the problem in this paper is how to how to develop a model of Fractional Integrated Recurrent Neural Networks (FIRNN) in forecasting time series data on National Electricity Load and how are the forecasting results of time series data on national electricity load using Fractional Integrated Recurrent Neural Networks (FIRNN). Furthermore, RNN models in long memory nonlinear models in this study will be called Fractional Integrated Recurrent Neural Networks (FIRNN). This research was conducted with literature studies, simulations and applications to real cases in memory of long time series data, by taking the case of the burden of the use of electricity in Indonesia. The previous studies show that most of the time series on consumption patterns of electrical load in Semarang city shows the pattern of long memory, because it has a fractional difference parameter which can be seen in [26].

Ranked Clusterability Model of Dyadic Data in Social Network

The dyads relationship as a substantial portion of triads or larger structure formed a ranked clusterability model in social network. Ranked clusterability model of dyads postulates that the hierarchical clustering process starts from the mutual dyads which occur only within clusters then stop until all of the mutual dyads grouped. The hierarchy process continues to cluster the asymmetric dyads which occur between clusters but at different levels. Then the last process is clustering the null dyads, which is clustered at the end of the hierarchy after all of asymmetric dyads grouped and occur only between clusters at the same level of the hierarchy. This paper explores a ranked clusterability model of dyads from a simple example of social network and represents it to the new sociomatrix that facilitate to view a whole network and presents the result in a dendrogram network data. This model adds a new insight to the development of science in a clustering study of emerging social network.

The Concept of Indiscernibility Level of Rough Set to Reduce the Dendrogram Instability

The main concept of rough sets theory is clustering similarities of objects based on the notions of indiscernibility relation. In this paper, we develop the concept of indiscernibility level of rough set theory as an additional measurement for hierarchical clustering. The combination between indiscernibility (quantitative indiscernibility relation) and indiscernibility level are used as a new method for hierarchical clustering. The indiscernibility level quantifies the indiscernibility of pairs of objects among other objects in information system. For comparison, the following four clustering methods were selected and evaluated on a simulation data set : average-, complete- and single-linkage agglomerative hierarchical clustering and Ward’s method. The simulation shows that the hierarchical clustering yields dendrogram instability that gives different solutions under permutations of input order of data objects. The result of this paper shows that the new method plays an important role in clustering information system and compared to other method, clustering based on indiscernibility and its indiscernibility level reduces the dendrogram instability.