Vol 53, No 1 (2019)

Fuzzy models have excellent capability of describing nonlinear system, so that a wide variety of models is proposed. However, these models are difficult to train and interpret with a hypothesis of data independence. Base on the above problems, an improved generalized fuzzy model is proposed by rules centralization, a new mixture model is built by substituting Epanechnikov quadratic kernel for Gaussian kernel of Gaussian mixture model, and the mutual transformation between the two models is proved. The proposed fuzzy model can not only provide good interpretability, but also easily estimate the parameters of the fuzzy model using the proposed mixture model. In addition, this proposed fuzzy model has higher segmentation efficiency for the input space with consideration of data correlation. The experimental results of a well-known nonlinear system identification show that the proposed fuzzy model has the best performance with fewer fuzzy rules and better generalization ability than some popular models.

The paper presents a methodology for bioinformatics data analysis. First, it describes the use of data analysis in bioinformatics—data preprocessing approaches, missing data processing approaches, data dimensionality reduction and classification algorithms. Then, the next section determines the most appropriate data analysis methods, which should be used in bioinformatics data analysis methodology to solve diagnostic classification task. The methodology was practically approbated in experiments using WEKA software and real-world bioinformatics data sets. This allowed determination of specific method realizations that show the best classification result; all intermediate results are recorded. Finally, the best preprocessing method sequence for this methodology is determined.

The optimization of task scheduling process in the cloud-computing environment is the multi-criteria NP-hard problem. The paper proposes a PSO based αPSO-TBLB (Task Based Load Balancing) load balancing method. The method provides an optimal migration of tasks causing overload from loaded virtual machines to corresponding virtual machines in the cloud environment. The minimization of task execution and transfer time in the suggested optimization model are chosen as target functions. The experimental testing of the suggested approach is carried out in Cloudsim and Jswarm software tools. As a result of the simulation based on proposed method found, the optimal solution for the scheduling of tasks and equal distribution of tasks to VMs (Virtual Machines) has been provided, and less time consumption has been achieved for the assignment process of tasks to VMs.

In this paper, we discuss data models and data mining for bicycles in Smart Cities. Mobility issues (or Smart Mobility) are one of the main components of Smart Cities. Bicycles, as a transport component in the cities, are on the rise all over the world. At least, it is true for all areas where the climate even minimally allows it. The reasons are quite obvious. This is democratic and accessible this type of transport, it is cheap and environmental friendliness. Of course, the promotion of a healthy lifestyle also plays its role. The development of this type of transport (like any other) has many different aspects. In this paper, we dwell on the issues of tracking the movement of cyclists and planning bike-sharing systems. All this information will serve as a set of metrics for any design in Smart Cities.

This paper studies observability and stabilizability of linear time-invariant systems in the presence of limited feedback data rates. The well-known data-rate theorem in the literature presents a lower bound on data rates, above which there exists a quantization, coding, and control scheme such that an unstable dynamical system can be stabilized. However, it is unnecessary to transmit data packet on the plant state to the controller when the state prediction error is small enough. Thus, we reduce the conservatism of the prior results by employing a time-varying coding scheme. It is shown in our results that, there exists the lower bound on the data rate for observability and stabilizability, which is tighter than the ones in the data-rate theorem in the literature. Illustrative examples are given to demonstrate the effectiveness of the proposed quantization, coding, and control scheme.