Search all articles

we propose a hybrid decision model for supporting the ranking financial status of corporations using case-based reasoning augmented with genetic algorithms and the fuzzy nearest neighbor method. An empirical experimentation on 746 cases was conducted that shows that the average accuracy of the ranking...

The main objective of this study is developing a linguistic cluster model in order to meet the public security index requirement and extract crime rule in time series. In contrast to the current studies in crime theory which mostly rely on traditional behavior science, we turned to a hybrid approach...

As a new general machine-learning tool based on structural risk minimization principle, Support Vector Machines (SVM) has the advantageous characteristic of good generalization. For this reason, the application of SVM in fault diagnosis field has becomes one growing reach focus. In this paper, data fusion...

In this paper we present a novel dense matching algorithm that relies on sparse stereo data in order to build a dense disparity map. The algorithm uses a recursive updating scheme to estimate the dense stereo data using various interpolation techniques. The major problem of classical template matching...

Fuzzy controllers in general are designed using the experts’ knowledge or human experience in order to decide the settings of a number of system factors. However, the relations between the factors and system performance are not easy to be determined. Usually, more than one system performance characteristic...

The forum system is useful for knowledge sharing and help-seeking. However, the existed forums are having the same problem such like nobody to answer the given questions. In order to attack the problem, this study proposes a human-expert oriented forum architecture to perform efficient knowledge sharing....

In this paper, we issue a hybrid system, which is integrated with the fuzzy system and Petri net models. We exploit the characteristics, imprecise or ambiguous information, of fuzzy theory to map the algorithm of crisp inputs and outputs, furthermore, model problems, product a nonlinear function, and...

An image-based fire detection method using neural networks is proposed in this paper. First, flame color features, based on the HSI color model, are trained by a backpropagation neural network for flame recognition. Then, based on the learned flame color features, regions with fire-like colors are roughly...

In this paper, an a-Si:H/a-SiC:H separated absorption and multiplication region superlattice avalanche photodiode (SAM-SAPD) is fabricated successfully by using combined effect of band edge discontinuity and build-in potential in p-n junction. We discuss the relationships among excess noise factor (Fe),...

This paper presents state-of-the-art silicon-based bandstop filters for the realization of microwave system-on-a-chip (SOC) communication systems. The characteristics of the bandstop filters are designed by the adjustment of the geometric dimensions of the filters in order to meet different system requirements....

We are developing an artificial stock market model with investor agents in order to analyze characteristics of a market price. In this paper, we analyze the effects of agent’s reaction rate to technical indicators in a financial market. Intensive experiments have suggested that trend follow investor...

Some approaches such as path indexing, labeling, and numbering scheme have been proposed in order to facilitate query of XML data. The indexes derived by these approaches must be rebuilt if XML data is updated. The method LSDX (Labeling Scheme for Dynamic XML data) can dynamically add new labels without...

In order to reduce the bias of linguistic recognitions among the group decision makings, this study develops “fuzzy multiple preference model” with two stages to better determining the suitable service innovation. A real numerical application in the chain wholesale has also been demonstrated.