Noriyuki Iwane

Constrained optimization by the ε constrained hybrid algorithm of particle swarm optimization and genetic algorithm

The e constrained method is an algorithm transformation method, which can convert algorithms for unconstrained problems to algorithms for constrained problems using the e level comparison that compares search points based on the constraint violation of them. We proposed the e constrained particle swarm optimizer ePSO, which is the combination of the e constrained method and particle swarm optimization. The ePSO can run very fast and find very high quality solutions, but the ePSO is not very stable and sometimes can only find lower quality solutions. On the contrary, the eGA, which is the combination of the e constrained method and GA, is very stable and can find high quality solutions, but it is difficult for the eGA to find higher quality solutions than the ePSO. In this study, we propose the hybrid algorithm of the ePSO and the eGA to find very high quality solutions stably. The effectiveness of the hybrid algorithm is shown by comparing it with various methods on well known nonlinear constrained problems.

Solving Nonlinear Constrained Optimization Problems by the ε Constrained Differential Evolution

The ε constrained method is an algorithm transformation method, which can convert algorithms for unconstrained problems to algorithms for constrained problems using the ε level comparison that compares search points based on the constraint violation of them. We propose the ε constrained differential evolution εDE, which is the combination of the ε constrained method and differential evolution (DE). DE is a simple, fast and stable search algorithm that is robust to multi-modal problems. It is expected that the εDE is robust to multi-modal problems, can run very fast and can find very high quality solutions. The effectiveness of the εDE is shown by comparing it with various methods on well known nonlinear constrained problems.

The integrated software product line model

In the industry, how to acquire customers needs and how to reflect them quickly into the software products are important issues. Recently, several architecture analysis and design methodologies have been developed and evaluated, and the software product line is receiving increasing attention for cost reduction in the software development. Combining the architectural requirements with the software product line systems is at the very center of what the software industry has to do in todays fast-changing environment. We developed the toolkit which automatically generates software programs from specifications. The toolkit was extended to the software product line system, and connected to the questionnaire system. This paper describes an integrated software product line model. A KITS (knowledge based integrated questionnaire-software product line) model which integrates software product line systems with questionnaire systems through a knowledge base system is proposed. It produce the efficient customized software systems while offering a substantial savings in time and cost.

VR-based dynamics learning system using haptic device and its evaluation

The role of computer tools and facilities become quite important to support educational activity in the school. In this view, many e-learning software are investigated, such as CAI (computer assisted instruction), WBT (Web based training), ILE (interactive learning environment), micro-world, CSCL (computer supported collaborative learning) (Galcev et al., 2001) and so on conventionally. On the other hand, VR (virtual reality) technology is focused as a new multimedia paradigm and is introduced into a lot of application software. And the costs of those VR facilities are reduced recently. For instance, a haptic device which offers the user such as force feedback would be powerful support tool for education at school and gives the student a direct feeling and more real feeling. In this paper, the new e-learning framework which is combined micro-world style ILE and VR concept is proposed. Also the prototype system which treats physical knowledge as the learning domain and introduces the PHANToM (personal haptic interface mechanism) as a haptic device is described.

Design based learning by knowledge reuse: Towards its application to e-learning

In order to support self-directed or subjective learning it is important that learners own knowledge are acquired and utilized in the learning process. It means that teaching material added by acquired knowledge should be different from each other. The learning process based on the teaching material is very personal and also a self-check for learning results is personal. Therefore it is necessary but difficult to support personal assessment manually because the self check or test for respective learners is different. To realize such a support system we propose a framework of design based learning by knowledge reuse and a method for implementing with an e-learning system. The essential and technical elements are made clear and our approach is explained from the feasibility and the reusability.

Question Generation for Learner Centered Learning

In the next generation of learning environment, a tablet PC with a stylus pen can be a learning tool. In this paper, a framework for learner centered question generation on the tablet PC is proposed. The framework is based on a learners actions and context of an expository text. In assisting active learning, three different types of questions can be generated according to the learners marking actions. Some associated questions can also be generated based on the type of a question. The generated questions will promote active learning by motivating further self-questionings and markings.

Structural learning of RBF-fuzzy rule bases based on information criteria and degeneration

There are some difficulties in researches on supervised learning using fuzzy rule-based systems: the difficulty of selecting effective input variables and the difficulty of selecting a proper rule structure. To solve these difficulties, we proposed GA/sup d/ (Genetic Algorithm with Degeneration), which performs structural learning. It is thought that if the information criteria of systems can be optimized, the obtained systems become more general, or can explain unlearned data better. But when GA/sup d/ optimizes the information criteria directly, there are some cases when too many parameters will be lost in earlier generations. In this study, the idea of multi-objective optimization, in which estimation errors and information criteria are both optimized, is introduced to avoid this problem. The weight for the errors is bigger in earlier generations and the weight for the information criteria becomes bigger in later generations. It is shown that this idea is effective to the structural learning of RBF fuzzy rule-based systems.

An approach to the software product line system for web applications

Quality, cost and delivery are the most crucial factors to be managed in developing software systems in the industry. The management of these software systems includes not only resolving the necessarily technical problems but also the organizational problems and the business problems. Companies depending upon their environments decide their own strategies to solve these problems. Software product line system is one of the solutions to these problems. This paper describes the way we experimented in these several years to develop the software systems at the company to meet some software development solutions. The efforts of Web-based application developments by the source code generator we developed are evaluated based on COCOMO model. Furthermore, the software development model for software product line systems using the toolkit is examined. This paper describes an approach to the software product line systems for Web application systems.

A social network model for big data privacy preserving and accountability assurance

Social networks have become valuable resources of big data exploration for individual users as well as for researchers and commercial organizations. However, how to protect user privacy while assure data credibility at the same time is still a big challenge both in practice and in academic research. This study presents a social network model for the purpose of big data privacy preserving and credibility assurance. In this paper, we discuss the mechanism of the model and present practical implementation designs on achieving the goals. Applications of the model in healthcare social network are also discussed.

Developing a Multi-Trust Model for Multi-Purpose Healthcare Social Networks

Online E-health platforms have enabled patients to obtain useful references on the Internet, as well as to share information and exchange experiences among people with similar diseases. Such social networks also enable researchers and corporates to conduct studies or advocates on targeted groups of patients. To make such a multi-purpose healthcare social network to be practical, concerns must be addressed regarding data security, patients privacy, user and data credibility, as well as how to promote user participation. This paper addresses these concerns as trust-engagement and develops a multi-trust model integrating public key encryption, access control, credible anonymity, and mutual rating mechanism for a multi-purpose healthcare social network.