Congestion Control Using Soft Computing

In the recent Internet era the queue management in the routers plays a vital role in the provision of Quality of Service (QoS). Virtual queue-based marking schemes have been recently proposed for Active Queue Management (AQM) in Internet routers. In this chapter, the authors propose Fuzzy enabled AQM (F-AQM) scheme where the linguistics variables are used to specify the behavior of the queues in the routers. The status of the queue is continuously monitored and decisions are made adaptively to drop or mark the packets as is done in Random Early Discard (RED) and Random Early Marking (REM) algorthms or schemes. The authors design a fuzzy rule base represented in the form of matrix indexed by queue length and rate of change of queue. The performance of the proposed F-AQM scheme is compared with several well-known AQM schemes such as RED, REM and Adaptive Virtual Queue (AVQ).

Interconnecting a Class of Machine Learning Algorithms with Logical Commonsense Reasoning Operations

The purpose of this chapter is to demonstrate the possibility of transforming a large class of machine learning algorithms into commonsense reasoning processes based on using well-known deduction and induction logical rules. The concept of a good classification (diagnostic) test for a given set of positive examples lies in the basis of our approach to the machine learning problems. The task of inferring all good diagnostic tests is formulated as searching the best approximations of a given classification (a partitioning) on a given set of examples. The lattice theory is used as a mathematical language for constructing good classification tests. The algorithms of good tests inference are decomposed into subtasks and operations that are in accordance with main human commonsense reasoning rules.

Type-2 Fuzzy Interface for Artificial Neural Network

Artificial Neural Network (ANN) based systems are bio-inspired mechanisms for intelligent decision support with capabilities to learn generalized knowledge from the large amount of data and offers high degree of self-learning. However, the knowledge in such ANN system is stored in the generalized connection between neurons in implicit fashion, which does not help in providing proper explanation and reasoning to users of the system and results in low level of user friendliness. On the other hand, fuzzy systems are very user friendly, represent knowledge in highly readable form and provide friendly justification to users as knowledge is stored explicitly in the system. Type-2 fuzzy systems are one step ahead while computing with words in comparison to typical fuzzy systems. This chapter introduces a generic framework of type-2 fuzzy interface to an ANN system for course selection process. Resulting neuro-fuzzy system offers advantages of self-learning and implicit knowledge representation along with the utmost user friendliness and explicit justification.

A Quick Presentation of Evolutionary Computation

Evolutionary computation is an old field of computer science, that started in the 1960s nearly simultaneously in different parts of the world. It is an optimization technique that mimics the principles of Darwinian evolution in order to find good solutions to intractable problems faster than a random search. Artificial Evolution is only one among many stochastic optimization methods, but recently developed hardware (General Purpose Graphic Processing Units or GPGPU) gives it a tremendous edge over all the other algorithms, because its inherently parallel nature can directly benefit from the difficult to use Single Instruction Multiple Data parallel architecture of these cheap, yet very powerful cards.

A Human–Machine Interface Design to Control an Intelligent Rehabilitation Robot System

The demand for rehabilitation increases daily as a result of diseases, occupational and traffic accidents and population growth. In the present time, some important problems occur regarding the rehabilitation period: the transportation of patients, the acquisition and storage of treatment data and the need to support the physiotherapists with intelligent devices. In order to overcome these challenges, the authors hereby propose a human machine interface to control an intelligent rehabilitation robot system designed for the lower limbs. The human machine interface has a structure that is created with a rule-based intelligent controlling structure, combined with conventional controller and an easy-to-use graphical user interface. By means of this interface, the rehabilitation sessions can be stored and members of the rehabilitation team can reach to this stored data via internet. Additionally, the patient can receive treatment in his house. One physiotherapist is able to treat several patients at a time by utilizing this system. The system’s capacity has been elaborated through the test results.

An Exposition of Feature Selection and Variable Precision Rough Set Analysis

This chapter considers, and elucidates, the general methodology of rough set theory (RST), a nascent approach to rule based classification associated with soft computing. There are two parts of the elucidation undertaken in this chapter, firstly the levels of possible pre-processing necessary when undertaking an RST based analysis, and secondly the presentation of an analysis using variable precision rough sets (VPRS), a development on the original RST that allows for misclassification to exist in the constructed “if … then …” decision rules. Throughout the chapter, bespoke software underpins the pre-processing and VPRS analysis undertaken, including screenshots of its output. The problem of US bank credit ratings allows the pertinent demonstration of the soft computing approaches described throughout.

A Bayesian Belief Network Methodology for Modeling Social Systems in Virtual Communities

Bayesian Belief Networks (BBNs) are increasingly used for understanding and simulating computational models in many domains. Though BBN techniques are elegant ways of capturing uncertainties, knowledge engineering effort required to create and initialize the network has prevented many researchers from using them. Even though the structure of the network and its conditional & initial probabilities could be learned from data, data is not always available and/or too costly to obtain. Further, current algorithms that can be used to learn relationships among variables, initial and conditional probabilities from data are often complex and cumbersome to employ. Qualitative-based approaches applied to the creation of graphical models can be used to create initial computational models that can help researchers analyze complex problems and provide guidance/support for decision-making. Once created, initial BBN models can be refined once appropriate data is obtained. This chapter extends the use of BBNs to help experts make sense of complex social systems (e.g., social capital in virtual communities) using a Bayesian model as an interactive simulation tool. Scenarios are used to update the model and to find out whether the model is consistent with the expert’s beliefs. A sensitivity analysis was conducted to help explain how the model reacted to different sets of evidence. Currently, we are in the process of refining the initial probability values presented in the model using empirical data and developing more authentic scenarios to further validate the model. We will elaborate on how database technologies were used to support the current approach and will describe opportunities for future database tools needed to support this type of work.

Fuzzy Decision Rule Construction Using Fuzzy Decision Trees

This chapter considers the soft computing approach called fuzzy decision trees (FDT), a form of classification analysis. The consideration of decision tree analysis in a fuzzy environment brings further interpretability and readability to the constructed ‘if .. then ..’ decision rules. Two sets of FDT analyses are presented, the first on a small example data set, offering a tutorial on the rudiments of one FDT technique. The second FDT analysis considers the investigation of an e-learning database, and the elucidation of the relationship between weekly online activity of students and their final mark on a university course module. Emphasis throughout the chapter is on the visualization of results, including the fuzzification of weekly online activity levels of students and overall performance.

A Combined GA-Fuzzy Classification System for Mining Gene Expression Databases

Microarray studies and gene expression analysis have received significant attention over the last few years and provide many promising avenues towards the understanding of fundamental questions in biology and medicine. In this chapter, the authors show that a combined GA-fuzzy classification system can be employed for effective mining of gene expression data. The applied classifier consists of a set of fuzzy if-then rules that allow for accurate non-linear classification of input patterns. A small number of fuzzy if-then rules are selected through means of a genetic algorithm, and are capable of providing a compact classifier for gene expression analysis. Experimental results on various well-known gene expression datasets confirm good classification performance of our approach.

Soft Computing Techniques in Spatial Databases

Spatial database systems and geographical information systems are currently only able to support geographical applications that deal with only crisp spatial objects, that is, objects whose extent, shape, and boundary are precisely determined. Examples are land parcels, school districts, and state territories. However, many new, emerging applications are interested in modeling and processing geographic data that are inherently characterized by spatial vagueness or spatial indeterminacy. Examples are air polluted areas, temperature zones, and lakes. These applications require novel concepts due to the lack of adequate approaches and systems. In this chapter, the authors show how soft computing techniques can provide a solution to this problem. They give an overview of two type systems or algebras that can be integrated into database systems and utilized for the modeling and handling of spatial vagueness. The first type system, called Vague Spatial Algebra (VASA), is based on well known, general, and exact models of crisp spatial data types and introduces vague points, vague lines, and vague regions. This enables an exact definition of the vague spatial data model since we can build it upon an already existing theory of spatial data types. The second type system, called Fuzzy Spatial Algebra (FUSA), leverages fuzzy set theory and fuzzy topology and introduces novel fuzzy spatial data types for fuzzy points, fuzzy lines, and fuzzy regions. This enables an even more fine-grained modeling of spatial objects that do not have sharp boundaries and interiors or whose boundaries and interiors cannot be precisely determined. This chapter provides a formal definition of the structure and semantics of both type systems. Further, the authors introduce spatial set operations for both algebras and obtain vague and fuzzy versions of geometric intersection, union, and difference. Finally, they describe how these data types can be embedded into extensible databases and show some example queries.