Fault-Tolerant Control of Mechanical Systems Using Neural Networks

Due to harsh working environment, control systems may degrade to an unacceptable level, causing more regular fault occurrences. In this case, it is necessary to provide the fault-tolerant control for operating the system continuously. The existing control techniques have given some ways to solve this problem, but if the system behaves in an unanticipated manner, then the control system may need to be modified, so that it handles the modified system. In this chapter, the authors are concerned with how this control system can be done automatically, and when it can be done successfully. They aimed in this work at handling unanticipated failure modes, for which solutions have not been solved completely. The model-based fault-tolerant controller with a self-detecting algorithm is proposed. Here, the radial basis function neural network is used in the controller to estimate the unknown failures. Once the failure is detected, the re-configured control is activated and then maintains the system continously. The fault-tolerant control is illustrated in two cases. It is shown that the proposed method can cope with different failure modes which are unknown a priori. The result indicates that the solution is suitable for a class of mechanical systems whose dynamics are subject to sudden changes resulting from component failures when working in a harsh environment.

Decay Detection in Citrus Fruits Using Hyperspectral Computer Vision

The citrus industry is nowadays an important part of the Spanish agricultural sector. One of the main problems present in the citrus industry is decay caused by Penicillium digitatum and Penicillium italicum fungi. Early detection of decay produced by fungi in citrus is especially important for the citrus industry of distribution. This chapter presents a hyperspectral computer vision system and a set of machine learning techniques in order to detect decay caused by Penicillium digitatum and Penicillium italicum fungi that produce more economic losses to the sector. More specifically, the authors employ a hyperspectral system and artificial neural networks. Nowadays, inspection and removal of damaged citrus is done manually by workers using dangerous ultraviolet light. The proposed system constitutes a feasible and implementable solution for the citrus industry; this has been proven by the fact that several machinery enterprises have shown their interest in the implementation and patent of the system.

Supervised Classification with Bayesian Networks

Bayesian Network classifiers (BNCs) are Bayesian Network (BN) models specifically tailored for classification tasks. There is a wide range of existing models that vary in complexity and efficiency. All of them have in common the ability to deal with uncertainty in a very natural way, at the same time providing a descriptive environment. In this chapter, the authors focus on the family of semi-naïve Bayesian classifiers (naïve Bayes, AODE, TAN, kDB, etc.), motivated by the good trade-off between efficiency and performance they provide. The domain of the BNs is generally of discrete nature, but since the presence of continuous variables is very common, the chapter discusses more classical and novel approaches to handling numeric data. In this chapter the authors also discuss more recent techniques such as multi-dimensional and dynamic models. Last but not least, they focus on applications and recent developments, including some of the BNCs approaches to the multi-class problem together with other traditionally successful and cutting edge cases regarding real-world applications.

Landmark Sliding for 3D Shape Correspondence

Shape correspondence, which aims at accurately identifying corresponding landmarks from a given population of shape instances, is a very challenging step in constructing a statistical shape model such as the Point Distribution Model. Many shape correspondence methods are primarily focused on closed-surface shape correspondence. The authors of this chapter discuss the 3D Landmark Sliding method of shape correspondence, which is able to identify accurately corresponding landmarks on 3D closed-surfaces and open-surfaces (Dalal 2007, 2009). In particular, they introduce a shape correspondence measure based on Thin-plate splines and the concept of explicit topology consistency on the identified landmarks to ensure that they form a simple, consistent triangle mesh to more accurately model the correspondence of the underlying continuous shape instances. The authors also discuss issues such as correspondence of boundary landmarks for open-surface shapes and different strategies to obtain an initial estimate of correspondence before performing landmark sliding.

Individual Prediction Reliability Estimates in Classification and Regression

Current machine learning algorithms perform well in many problem domains, but in risk-sensitive decision making – for example, in medicine and finance – experts do not rely on common evaluation methods that provide overall assessments of models because such techniques do not provide any information about single predictions. This chapter summarizes the research areas that have motivated the development of various approaches to individual prediction reliability. Based on these motivations, the authors describe six approaches to reliability estimation: inverse transduction, local sensitivity analysis, bagging variance, local cross-validation, local error modelling, and density-based estimation. Empirical evaluation of the benchmark datasets provides promising results, especially for use with decision and regression trees. The testing results also reveal that the reliability estimators exhibit different performance levels when used with different models and in different domains. The authors show the usefulness of individual prediction reliability estimates in attempts to predict breast cancer recurrence. In this context, estimating prediction reliability for individual predictions is of crucial importance for physicians seeking to validate predictions derived using classification and regression models.

Automatic Text Classification from Labeled and Unlabeled Data

Automatic text classification is a process that applies information retrieval technology and machine learning algorithms to build models from pre-labeled training samples and then deploys the models to previously unseen documents for classification. Text classification has been widely applied in many fields ranging from Web page indexing, document filtering, and information security, to business intelligence mining. This chapter presents a semi-supervised text classification framework that is based on the radial basis function (RBF) neural networks. The framework integrates an Expectation Maximization (EM) process into a RBF network and can learn for classification effectively from a very small quantity of labeled training samples and a large pool of additional unlabeled documents. The effectiveness of the framework is demonstrated and confirmed by some experiments of the framework on two popular text classification corpora.

Supervision of Industrial Processes using Self Organizing Maps

The objective of this chapter is to present, in a comprehensive and unified way, a corpus of data and knowledge visualization techniques based on the Self-Organizing Map (SOM). These techniques allow exploring the behavior of the process in a visual and intuitive way through the integration of existing process-related knowledge with information extracted from data, providing new ways for knowledge discovery. With a special focus on the application to process supervision and modeling, the chapter reviews well known techniques –such as component planes, u-matrix, and projection of the process state– but also presents recent developments for visualizing process-related knowledge, such as fuzzy maps, local correlation maps, and model maps. It also introduces the maps of dynamics, which allow users to visualize the dynamical behavior of the process on a local model basis, in a seamless integration with the former visualizations, making it possible to confront all them for discovery of new knowledge.

Probabilistic Graphical Models for Sports Video Mining

Semantic analysis is an active and interesting research topic in the field of sports video mining. In this chapter, the authors present a multi-level video semantic analysis framework that is featured by hybrid generative-discriminative probabilistic graphical models. A three-layer semantic space is proposed, by which the semantic video analysis is cast into two inter-related inference problems defined at different semantic levels. In the first stage, a multi-channel segmental hidden Markov model (MCSHMM) is developed to jointly detect multiple co-existent mid-level keywords from low-level visual features, which can serve as building blocks for high-level semantics. In the second stage, authors propose the auxiliary segmentation conditional random fields (ASCRFs) to discover the game flow from multi-channel key-words, which provides a unified semantic representation for both event and structure analysis. The use of hybrid generative-discriminative approaches in two different stages is proved to be effective and appropriate for multi-level semantic analysis in sports video. The experimental results from a set of American football video data demonstrate that the proposed framework offers superior results compared with other traditional machine learning-based video mining approaches.

Agent Based Systems to Implement Natural Interfaces for CAD Applications

Currently, important advances are being carried out in CAD (Computer Aided Design) applications; however, these advances have not yet taken place for CAS (Computer Aided Sketching) applications. These applications are intended to replace complex menus with natural interfaces that support sketching for commands and drawing, but the recognition process is very complex and doesn’t allow its application yet. So, although natural interfaces for CAD applications have not yet been solved, works based on sketching devices have been explored to some extent. In this work, the authors propose a solution for the problem of recognition of sketches using an agent-based architecture, which distributes the agents hierarchically to achieve the best decision possible and to avoid reliance on of the drawing sequence.

Learning and Explaining the Impact of Enterprises’ Organizational Quality on their Economic Results

The authors employed traditional and novel machine learning to improve insight into the connections between the quality of an organization of enterprises as a type of formal social units and the results of enterprises’ performance in this chapter. The analyzed data set contains 72 Slovenian enterprises’ economic results across four years and indicators of their organizational quality. The authors hypothesize that a causal relationship exists between the latter and the former. In the first part of a two-part process, they use several classification algorithms to study these relationships and to evaluate how accurately they predict the target economic results. However, the most successful models were often very complex and difficult to interpret, especially for non-technical users. Therefore, in the second part, the authors take advantage of a novel general explanation method that can be used to explain the influence of individual features on the model’s prediction. Results show that traditional machine-learning approaches are successful at modeling the dependency relationship. Furthermore, the explanation of the influence of the input features on the predicted economic results provides insights that have a meaningful economic interpretation.