Parallelized fusion visualization system

The speed-up of supercomputers has increased the complexity of simulations. To analyze such kind of data, we believe that new types of visualization software are needed. Therefore, we have been developing a visualization system called “Fusion Visualization”, and the progresses were reported in the AROB 18th and 19th International Symposiums. We introduced the overall concept at the AROB 18th International Symposium, and then demonstrated a sample of flow visualization in a blood vessel in the AROB 19th International Symposium. To extend our system to enable the handling of larger data, we have implemented the proposed system on a parallelized visualization system; AVS/Express PCE (Parallel Cluster Edition). This paper describes the implementation and the benchmark results.     

New time-varying fuzzy sets based on a PSO midpoint of the universe of discourse

The paper presents a robust parallel distributed compensation(PDC) fuzzy controller for a nonlinear and certain system in continuous time described by the Takagi-Sugeno(T-S) fuzzy model. This controller is based on a new type of time-varying fuzzy sets(TVFS). These fuzzy sets are characterized by displacement of the kernels to the right or left of the universe of discourse, and they are directed by a well-defined criterion. In this work, we only focused on the movement of midpoint of the universe. The movements of this midpoint are optimized by particle swarm optimization(PSO) approach.     

PID controller tuning for a multivariable glass furnace process by genetic algorithm

Standard genetic algorithms (SGAs) are investigated to optimise discrete-time proportional-integral-derivative (PID) controller parameters, by three tuning approaches, for a multivariable glass furnace process with loop interaction. Initially, standard genetic algorithms (SGAs) are used to identify control oriented models of the plant which are subsequently used for controller optimisation. An individual tuning approach without loop interaction is considered first to categorise the genetic operators, cost functions and improve searching boundaries to attain the desired performance criteria. The second tuning approach considers controller parameters optimisation with loop interaction and individual cost functions. While, the third tuning approach utilises a modified cost function which includes the total effect of both controlled variables, glass temperature and excess oxygen. This modified cost function is shown to exhibit improved control robustness and disturbance rejection under loop interaction.     

Fuzzy clustering with learnable cluster-dependent kernels

We propose a new relational clustering approach, called Fuzzy clustering with Learnable Cluster-dependent Kernels (FLeCK), that learns the underlying cluster-dependent dissimilarity measure while seeking compact clusters. The learned dissimilarity is based on a Gaussian kernel function with cluster-dependent parameters. Each cluster’s parameter learned by FLeCK reflects the relative intra-cluster and inter-cluster characteristics. These parameters are learned by optimizing both the intra-cluster and the inter-cluster distances. This optimization is achieved iteratively by dynamically updating the partition and the local kernel. This makes the kernel learning task takes advantages of the available unlabeled data and reciprocally, the categorization task takes advantages of the learned local kernels. Another key advantage of FLeCK is that it is formulated to work on relational data. This makes it applicable to data where objects cannot be represented by vectors or when clusters of similar objects cannot be represented efficiently by a single prototype. Using synthetic and real data sets, we show that FLeCK learns meaningful parameters and outperforms several other algorithms. In particular, we show that when data include clusters with various inter- and intra-cluster distances, learning cluster-dependent parameters is crucial in obtaining a good partition.     

Approximate variational inference based on a finite sample of Gaussian latent variables

Variational methods are employed in situations where exact Bayesian inference becomes intractable due to the difficulty in performing certain integrals. Typically, variational methods postulate a tractable posterior and formulate a lower bound on the desired integral to be approximated, e.g. marginal likelihood. The lower bound is then optimised with respect to its free parameters, the so-called variational parameters. However, this is not always possible as for certain integrals it is very challenging (or tedious) to come up with a suitable lower bound. Here, we propose a simple scheme that overcomes some of the awkward cases where the usual variational treatment becomes difficult. The scheme relies on a rewriting of the lower bound on the model log-likelihood. We demonstrate the proposed scheme on a number of synthetic and real examples, as well as on a real geophysical model for which the standard variational approaches are inapplicable.     

A large vocabulary system for Arabic online handwriting recognition

The success of using Hidden Markov Models (HMMs) for speech recognition application has motivated the adoption of these models for handwriting recognition especially the online handwriting that has large similarity with the speech signal as a sequential process. Some languages such as Arabic, Farsi and Urdo include large number of delayed strokes that are written above or below most letters and usually written delayed in time. These delayed strokes represent a modeling challenge for the conventional left-right HMM that is commonly used for Automatic Speech Recognition (ASR) systems. In this paper, we introduce a new approach for handling delayed strokes in Arabic online handwriting recognition using HMMs. We also show that several modeling approaches such as context based tri-grapheme models, speaker adaptive training and discriminative training that are currently used in most state-of-the-art ASR systems can provide similar performance improvement for Hand Writing Recognition (HWR) systems. Finally, we show that using a multi-pass decoder that use the computationally less expensive models in the early passes can provide an Arabic large vocabulary HWR system with practical decoding time. We evaluated the performance of our proposed Arabic HWR system using two databases of small and large lexicons. For the small lexicon data set, our system achieved competing results compared to the best reported state-of-the-art Arabic HWR systems. For the large lexicon, our system achieved promising results (accuracy and time) for a vocabulary size of 64k words with the possibility of adapting the models for specific writers to get even better results.     

The development of a new integral control model based on the analysis of three complex systems in Serbia

The aim of this paper is to analyse the results of research carried out in the railway, mining, and electric power systems in Serbia and form a new integral control model. The three methodological procedures are applied. First, analytical-synthetic methodological approach breaks down complex technical system into three parts: bio-cybernetic system, “operator”; technical system, “technology”; and additional system, “working environment.” Second, network planning method is used to analyse time, according to the critical path method. Third, fuzzy analytic hierarchy process determines the key research factors. General results of research are new integral control model, and new research areas and activities. The most prominent factors are: in “bio-cybernetic system”—operator’s arm reach, body postures and movement sequences, operator’s work, occurrence of stress, and occurrence of fatigue; in “technical system”—location and dimension of control desk, display panel, video display terminal, symbols on video display terminal, colours in control centres, and suitability of the keyboards; and in “supporting system”—illumination in control centres and relative humidity. Based on the analysis of factors and synthesis of results, the following recommendation are proposed: new control desk design; new display panel design; new design of the main and local lighting; new illumination and contrast characteristics, and environmental impact assessment. For research on a variety of complex technical systems, new integral control model can be applied, with corresponding extensions.     

Socially shared regulation of learning in CSCL: understanding and prompting individual- and group-level shared regulatory activities

The field of computer supported collaborative learning (CSCL) is progressing instrumentally and theoretically. Nevertheless, few studies examine the effectiveness and efficiency of CSCL with respect to cognitive, motivational, emotional, and social issues, despite the fact that the role of regulatory processes is critical for the quality of students’ engagement in collaborative learning settings. We review the four earlier lines in developing support in CSCL and show how there has been a lack of work to support individuals in groups to engage in, sustain, and productively regulate their own and the group’s collaborative processes. Our aim is to discuss how our conceptual work in socially shared regulation of learning (SSRL) contributes to effective and efficient CSCL, what tools are presently available, and what the implications of research on these tools are for future tool development.