Earticle

In this paper, we present a rule extraction method using a modified fuzzy min-max neural network for dynamic hand gesture recognition. We introduce a feature relevance measure for the pattern classification based on FMM neural networks. During the learning process, the feature distribution information is utilized to compensate the hyperbox distortion which may be caused by eliminating the overlapping area of hyperboxes in the contraction process. We define a feature saliency measure that represents a degree of relevance of a feature in a classification problem. From the measure, we can classify excitatory features and the inhibitory features which can be used for the rule generation process.

This paper describes an agent-oriented requirements engineering approach. Agent-oriented techniques represent an exciting new means of analysing, designing and building complex software systems. They have the potential to significantly improve current practice in software engineering and to extend the range of applications that can feasibly be tackled. i * can be used to model social dependencies between agents and how process design choices affect the agents’ goals. Agent-oriented approaches are becoming popular in software engineering, both as architectural frameworks, and as modeling frameworks for requirements engineering and design. i* is an informal diagram-based language for early -phase requirements engineering that supports the modeling of social dependencies between agents and how process design choices affect the agents’ goals both functional and non-functional. ConGolog is an expressive logic-based formalism for specifying processes that involves multiple agents. The two formalisms complement each other well, and in this work, we used the combination of the two frameworks in requirements engineering. The i* SR-diagram language is extended with process specification annotations, which allow the SR model of a system to be refined and then mapped into a ConGolog model. The mapping must satisfy a set of mapping rules, which ensure that it specifies which elements in the two models are related and that the models are consistent.

We proposed a new non-uniform sampling and quantization using variables low-pass filter with correlation. It focuses on the naturalness and intelligibility of speech synthesis applications and the compression and signal-to-noise ratio of speech transmission applications. However, it is well known that when conventional sampling methods are applied directly to speech signal, the required amount of data is comparable to or more than that of uniform sampling method. To overcome this problem, a new non-uniform methods is proposed, in which time domain coding is applied to two low-pass filters in lower bandwidth and the remain signals are compensated by the Gaussian white signal, which is used to get high quality speech by correlation of signal .

RESTful web service developers to easily deploy their web service through HTTP protocol. We propose two solutions to make easier tax adjustment; tax overpayment and tax return expectation. First, even though almost all WTO countries‘ tax law explicitly states the tax incentive to subsidies for research and development, many companies often incorrectly classify the subsidies for research as a taxable income not as a non-taxable income, resulting in over payment of tax for taxpayers. Thus, we introduce a RESTful Open API for checking whether earned income year-end tax settlement is correct or not. This API automatically audits the tax settlement by some required numbers from the earned income withholding receipt. Second, If a taxpayer has wage & salary income only, he/she is not subject to final global income tax return because of the self-assessment. But, if he/she has other incomes, the tax return on those must be completed and submitted by the end of May of the next year. Second, we focus on tax return simulation by global income settlement. The reason why we are focusing onto this issues is because a large amount of the other income tax which was already deducted as withholding could be returned. Thus, we provide a REST Web Service Open API so that we can easily calculate how much we can get tax return by just invoking this interface through the Internet. Moreover we implemented a smartphone application by using HTML5 and sencha touch UI framework. The proposed approach has a lot of advantages : platform independence which is available on any platform such as android, iphone, and even web on windows machine, easy update which is adaptable to any change of tax law.

In the majority of papers on rough set theory itis assumed that the information is complete, i.e., that forall cases all attribute values and decision values arespecified. Such a decision table is said to be completelyspecified.In practice, however, input data, presented as decisiontables, may have missing attribute and decision values,i.e., decision tables are incompletely specified. In this paper we use a variation relation describing the decision table with missing attribute values, i.e. replacing all the missing attribute values by minimum error sums of square for the total variation and thereby completing the information table. Subsequently, we find the reduct and core of the complete decision table and verify that the reduct and core by our method is better than ones by ROSE2 software. Thereafter we generate the rules based on reduct. The most important thing is to be different in the decision rules according to handling missing attribute values.

An existing application software perhaps has no requirements document or requirements document does not represent the application. The situation creates a problem in application software maintenance or reengineering. Thus, requirements document should be reconstructed from the existing application software. Effort on reconstructing a requirements document from the existing application software is similar to requirements document construction for a new application software, needs requirements elicitation. Requirements elicitation can be done using an ontology model that captures end-to-end interaction between users and system. Using the ontology model does not need software documentation and source code at all, the only one is the existing application software itself. In this case, the existing application software should be executed and then observed its behavior in term of end-to-end interaction on each starting point in which the interface identified. This set of interaction flows is known as a Use Case Realization (UCR) in Requirements Management with Use Case (RMUC) approach. The appropriate ontology model provides an ease to understand the software requirements semantic meaning. The model consists of two models includes, WIMP-UI (Window Icon Menu Pointer – User Interface) and USI (User - Interaction) ontology models.

We present a stereoscopic rendering scheme of 3D models with artistic styles including cartoon rendering contour drawing. Our scheme design separate projection matrices for both eyes and apply them to generate two images. After that, we apply artistic styles for the images. To preserve the coherence of the images, we design an image space algorithm that modifies the artistic effects of both images to reduce the difference of the artistic styles.

Readability is one important quality attributes for software source codes. Readability has also significant relation or impact with other quality attributes such as: reusability, maintainability, reliability, complexity, and portability metrics. This research develops a novel approach called Impact of Programming Features on Code Readability (IPFCR), to examine the influence of various programming features and the effect of these features on code readability. A code Readability Tool (CRT) is developed to evaluate the IPFCR readability features or attributes. In order to assess the level if impact that each one of the 25 proposed readability features may have, positively or negatively on the overall code readability, a survey was distributed to a random number of expert programmers. These experts evaluated the effect of each feature on code readability, based on their knowledge or experience. Expert programmers have evaluated readability features to be ordered then classified into positive and negative factors based on their impact on code readability or understanding. The survey responses were analyzed using SPSS statistical tool. Most of proposed code features showed to have significantly positive impact on enhancing readability including: meaningful names, consistency, and comments. On the other hand, fewer features such as arithmetic formulas, nested loops, and recursive functions showed to have a negative impact. Finally, few features showed to have neutral impact on readability.

In this paper, two methods of extracting objects are compared through application to underwater images: one method is to extract objects by removing the background and quantifying it into a codebook by measuring the Mahalanobis distance for accurate object segmentation and extraction, and the other is to extract objects by removing the background and quantifying it into a codebook by measuring the Euclidean distance. In an experiment relating to the comparison and analysis, a standard underwater sample image was learned. Then, the background color’s average value and the input image’s stochastic distances were computed through the color similarity algorithm, and then the object was extracted after the background color could be removed. For the performance evaluation on the two algorithms, an underwater image was used to run some computer simulations. The experiment showed that an image applied with the color similarity algorithm had a better image segmentation performance than that with the different image technique.