Catálogo de publicaciones - libros

We apply automated classification techniques to determine whether dementia is associated with changes in the physical structure of small blood vessels in the brain. A successful predictive model would imply such an association. The use of measures derived from fractal analysis, and the use of machine learning classification algorithms, allow exploration of highly complex relationships. Results suggest that although physiological differences are difficult to detect, and vary between different areas of brain tissue, there is evidence for such an association. If such changes can be detected from images of post mortem tissue, this implies that investigation of the medical significance of these changes could provide greater understanding of this class of diseases.

We consider the issue of model selection for some prediction problems in consumer finance. In particular, we look at performance metrics in the context of classification problems. Example areas considered include response modeling, profitability modeling and default prediction in the framework of a customer relationship management (CRM) system. We propose some guidelines for choosing the appropriate performance measure for the predictive model based on the decision framework it is part of.

This paper presents microcontroller based autotuning proportional-integral-derivative (PID) controller for an oven designed as an experiment set. Different types of autotuning PID controller methods have been examined. Proportional, P, control method has been applied first. Relay and integral square time error criterion (ISTE) tuning methods are used as autotuning PID method. For relay tuning method, proportional (P), proportional-integral (PI) and proportional-integral-derivative (PID) and for ISTE disturbance (PI, PID) have been used. These methods have been applied to the experiment set which is an FODPT (First Order Plus Dead Time) system. To be able to control this system a digital signal processing card is designed. PIC17C44 is used as microcontroller and ADS1212 is used as A/D converter. And the results are discussed to define which controller is the best for this experiment set.

We propose a new ε -ASA (Aggregation based on Situation Assessment) algorithm based on the fuzzy situation assessment model (FSAM) to reflect a situation in the process of aggregation. The proposed aggregation algorithm makes an adaptive aggregation result between min and max depending on the value of parameter representing a degree of situation. It is a further step toward situation-based aggregation.

This paper proposes a novel particle swarm optimization (PSO) for solving constrained optimization problems. Based upon the acceptable assumption that any feasible solution is better than any infeasible solution, a new mechanism for constraints handling is incorporated in the standard particle swarm optimization. In addition to the mechanism of constraints handling, a mutation strategy to increase population diversity is added to the proposed algorithm to improve convergence. Experimental results compared with genetic algorithm and a standard PSO show that the proposed algorithm is a desirable and competitive algorithm for solving constrained optimization problems.

Link discovery is an emerging research direction for extracting evidences and links from multiple data sources. This paper proposes a self-organizing framework for discovering links from multi-relational databases. It includes main functional modules for developing adaptive data transformers and representation specification, multi-relational feature construction, and self-organizing multi-relational correlation and link discovery algorithms.

This paper presents the Intelligent Power Quality Disturbance Analysis (IPQDA) software tool that is designed for an automatic analysis of power quality (PQ) disturbance. The main capabilities of the software include analysis of disturbance waveforms, identification of a particular type of disturbance and notification of a disturbance. Another important feature of the program is that it can automatically send email or short messaging notifications upon identification of a disturbance to alert the system operator of a disturbance.

Robots successfully manipulate objects in controlled environments. However, they fail in unknown environments. Few years old children lift and manipulate unfamiliar objects more dexterously than today’s robots. Therefore, roboticists are looking for inspiration on neurophysiological studies to improve their robotics control models. We present an artificial intelligence control model for dexterous manipulation, and a grip and load force control algorithm, strongly inspired on neurophysiological studies of the human manipulation process.

Synchronization in chaotic oscillatory systems has a wide array of applications in biology, physics and communication systems. Over the past 10 years there has been considerable interest in the synchronization properties of small-world and scale-free networks. In this paper, we define the fitness of a configuration of coupled oscillators as its ability to synchronize. We then employ an optimization algorithm to determine network structures that lead to an enhanced ability to synchronize. The optimized networks generally have low clustering, small diameters, short path-length, are disassortative, and have a high degree of homogeneity in their degree and load distributions.

To make full use of inside and outside resources in a competitive globalization market, many manufacturers and service providers are seeking a strategic cooperation with suitable vendors to improve their supply chain management (SCM) so that they can concentrate their efforts on their own core business. In this research, a comprehensive evaluation approach is presented for optimal combination selection among candidate vendors and outsourced parts. An evaluation system with a set of vendor selection indices is established. Also, a hierarchical fuzzy model for vendor selection is developed. And thus, a progressively-simplified approach is presented to deal with the vendor combination selection problem in a supply chain system. Finally, an example is given to illustrate the effectiveness of this approach.