Earticle

As the ubiquitous computing becomes popular, its applications come to real life as a form of a wide variety of ubiquitous decision support systems (UDSS). However, such ubiquity should be supported by prediction capability no matter which kind of contexts users are in. In this sense, context prediction capability, which is to predict future contexts users are going to enter sooner or later, becomes an extremely important part of ubiquitous decision support systems. This study proposes a new breed of context prediction mechanism using the Markov Blanket obtained from General Bayesian Network (GBN) as a main vehicle. To improve the prediction accuracy, ensemble of robust prediction classifiers is suggested on the basis of the GBN Markov Blanket. Three classifiers included in the ensemble mechanism are Bayesian networks, decision classifiers, and an SVM (Support Vector Machine). The proposed GBN Markov blanket-assisted ensemble classifier is applied to a real dataset of location prediction. Results were promising enough to conclude that the proposed ensemble classifier based on the GBN Markov Blanket is worthwhile for being adopted in developing a powerful context prediction purpose UDSS. Practical implications are also discussed with future research issues.

This paper proposed the U-bus system design based on wireless sensor network (WSN) for blind people. This system has two main parts. First part is blind people recognition. Another part is communication between a bus and bus station. Blind people recognition part is constructed simple device and system. This part decides existing or non-existing of the blind at bus station. And then if pre-process recognize blind people, the bus station will communicate the bus. We make up the announcement system about arrived bus information for the blind people using these parts. This announcement about arrived bus is very useful to blind people for taking the bus.

Ubiquitous decision support systems have remained an imaginary and almost useless system for decades since its first introduction in early 1990’. However, it came out of lab into real world as ubiquitous computing became tangible in the form of mobile devices, pervasive mechanisms, and various mobile Internet technologies. Typically, context-aware systems had received acclaims from both researchers and practitioners as an alternative to making ubiquitous systems touch-and-feel electronics to the users. Nevertheless, context-aware systems lack predictive power which is essential for any ubiquitous systems to suggest timely and effective information for users. Poorly predicted information is likely to degrade the ubiquitous systems seriously. In this respect, context prediction mechanism emerges as a reliable vehicle for making ubiquitous systems more sustainable decision support tool for users. Despite the potentials of context prediction mechanism, few reliable mechanisms exist in literature which shows robust performance against changes in user’ contexts. For this reason, we propose a new type of ubiquitous decision support system that is powered by General Bayesian Network (GBN) capable of organizing causal relationships among a set of related variables. Drawing on the GBN’ strengths, this study proposes U-BASE (Ubiquitous Bayesian network-Assisted Support Engine) to suggest more reliable solution for the context prediction tasks. Performance of U-BASE was tested against real contextual data set, garnering very robust results. The practical implications are fully discussed with some future research issues.

The population based approach of GA allows large jumps in the search space. However GAs has not proved successful for graph coloring because of the large degree of symmetry of the solution space. In fact, because of this symmetry mismatch it is very unlikely to produce a fit offspring when combining two good solutions. Thus GAs are often considered on inappropriate approach for problems such as graph coloring with a highly degenerate objective function. In order to compensate for this degeneracy we have applied Bitstream neuron (Boltzmann Machine) to the solution obtained from GA. Unlike traditional approaches of GA and ANN the proposed hybrid algorithm is guaranteed to have 100% convergence rate to valid solution with no parameter tuning. Experiments of such a hybrid algorithm are carried out on large DIMACS Challenge benchmark graphs. Results prove very competitive. Analysis of the behavior of the algorithm sheds light on ways to further improvement.