Inductive Learning Techniques Research Articles

Unsupervised, language-independent grapheme-to-phoneme conversion by latent analogy

Automatic, data-driven grapheme-to-phoneme conversion is a challenging but often necessary task. The top-down strategy implicitly followed by traditional inductive learning techniques tends to dismiss relevant contexts when they have been seen too infrequently in the training data. The bottom-up philosophy inherent in pronunciation by analogy allows for a markedly better handling of unusual patterns, but also relies heavily on individual, language-dependent alignments between letters and phonemes. To avoid such supervision, this paper proposes an alternative solution, dubbed pronunciation by latent analogy, which adopts a more global definition of analogous events. For each out-of-vocabulary word, a neighborhood of globally relevant pronunciations is constructed through an appropriate data-driven mapping of its graphemic form. Phoneme transcription then proceeds via locally optimal sequence alignment and maximum likelihood position scoring. This method was successfully applied to the speech synthesis of proper names with a large diversity of origin.

NEURO-FUZZY MODELLING AND CONTROL OF ROBOT MANIPULATORS FOR TRAJECTORY TRACKING

This paper presents a new neuro-fuzzy controller for robot manipulators. First, an inductive learning technique is applied to generate the required modelling rules from input/output measurements recorded in the off-line structure learning phase. Second, a fully differentiable fuzzy neural network is developed to construct the inverse dynamics part of the controller for the on-line parameter learning phase. Finally, a fuzzy-PID-like incremental controller was employed as feedback servo-controller. The proposed control system was tested using dynamic model of a six-axis industrial robot. The control system showed good results compared to the conventional-PID individual joint controller.

INTELLIGENT INTERNAL MODEL CONTROL OF ROBOTS FOR UPPER-LIMB REHABILITATION

This paper presents a neuro-fuzzy internal model Cartesian controller for robot manipulators. An inductive learning technique is applied to generate the required inverse dynamics and inverse kinematics modelling rules from input/output measurements. A fully differentiable fuzzy neural network is used to construct the adaptive sections of the controller for on-line parameters adaptation. A fuzzy-PID-like incremental controller is employed as feedback servo-controller. The internal model Cartesian controller is implemented using inverse kinematics and forward kinematics models of the robot. The proposed control system was tested using a dynamics model of a six-axis industrial robot to perform upper-limb rehabilitation. The obtained results demonstrate the validity of the proposed control scheme.

A rule merging technique for handling noise in inductive learning

Inductive learning algorithms are used for extracting IF-THEN rules from examples. The main weakness of most existing algorithms is their poor ability to handle data containing noise. This problem is even more severe when inductive learning techniques are applied to real engineering data. The paper presents a new pruning technique that improves significantly the performance of the RULES family of inductive learning algorithms. The technique is designed for RULES-5, the latest algorithm in the family, but could readily be applied to rule sets created by other algorithms.

Intelligent condition monitoring using fuzzy inductive learning

Extensive research has been performed for developing knowledge based intelligent monitoring systems for improving the reliability of manufacturing processes. Due to the high expense of obtaining knowledge from human experts, it is expected to develop new techniques to obtain the knowledge automatically from the collected data using data mining techniques. Inductive learning has become one of the widely used data mining methods for generating decision rules from data. In order to deal with the noise or uncertainties existing in the data collected in industrial processes and systems, this paper presents a new method using fuzzy logic techniques to improve the performance of the classical inductive learning approach. The proposed approach, in contrast to classical inductive learning method using hard cut point to discretize the continuous-valued attributes, uses soft discretization to enable the systems have less sensitivity to the uncertainties and noise. The effectiveness of the proposed approach has been illustrated in an application of monitoring the machining conditions in uncertain environment. Experimental results show that this new fuzzy inductive learning method gives improved accuracy compared with using classical inductive learning techniques.

Is inductive machine learning just another wild goose (or might it lay the golden egg)?

The research reported here contrasts the roles, methodologies and capabilities of statistical methods with those of inductive machine learning methods, as they are used inferentially in geographical analysis. To this end, various established problems with statistical inference applied in geographical settings are reviewed, based on Gould's (1970) critique. Possible solutions to the problems outlined by Gould are suggested via reviews of: (i) improved statistical methods, and (ii) recent inductive machine learning techniques. Following this, some newer problems with inference are described, emerging from the increased complexity of geographical datasets and from the analysis tasks to which we put them. Again, some solutions are suggested by pointing to newer methods. By way of results, questions are posed, and answered, relating to the changes brought about by adopting inductive machine learning methods for geographical analysis. Specifically, these questions relate to analysis capabilities, methodologies, the role of the geographer and consequences for teaching and learning. Conclusions argue that there is now a strong need, motivated from many perspectives, to give geographical data a stronger voice, thus favouring techniques that minimize the prior assumptions made of a dataset.

Prediction of Remaining Service Life of Bridge Decks Using Machine Learning

This paper demonstrates that it is feasible to use either the k-nearest-neighbor instance-based learning (IBL) technique or the inductive learning (IL) technique for engineering applications in classification/prediction problems such as estimating remaining service life of bridge decks. It is shown that IBL is more efficient than IL: The best achieved percentages of correctly classified instances are 50% as generated by k-nearest-neighbor IBL and 41.8% when generated by the C4.5/IL learning algorithm. From a machine learning (ML) standpoint both these values are considered low, but this is attributed to the fact that the deterioration model used to compute the remaining service life turned out to be inadequate. It is based on a methodology developed under the Strategic Highway Research Program (SHRP) for life-cost analysis of concrete bridges relative to reinforcement corrosion. Actual bridge deck surveys were obtained from the Kansas Department of Transportation that include the type of attributes needed for the SHRP methodology. The experimentation with the ML algorithms reported here also describes the experience one may go through when faced with an imperfect model, or with incomplete data or missing attributes.

An approach to learning from both good and poor factory performance in a Kanban-based just-in-time production system

In a JIT manufacturing environment it may be desirable to learn from an archived history of data that contains information that reflects less than optimal factory performance. The purpose of this paper is to use rule induction to predict JIT factory performance from past data that reflects both poor (saturated or starved) and good (efficient) factory performance. Inductive learning techniques have previously been applied to JIT production systems (Markham et al. , Computers and Industrial Engineering, 34 , 717-726, 1998; Markham et al. , International Journal of Manufacturing Technology Management, 11 (4), 239-246, 2000), but these techniques were only applied to data sets that reflected a well-performing factory. This paper presents an approach based on inductive learning in a JIT manufacturing environment that (1) accurately classifies and predicts factory performance based on shop factors, and (2) identifies the important relationships between the shop factors that determine factory performance. An example application is presented in which the classification and regression tree (CART) technique is used to predict saturated, starved or efficient factory performance based on dynamic shop floor data. This means that the relationship between the variables that cause poor factory performance can be discovered and measures to assure efficient performance can then be taken.

High-Performance Commercial Data Mining: A Multistrategy Machine Learning Application

We present an application of inductive concept learning and interactive visualization techniques to a large-scale commercial data mining project. This paper focuses on design and configuration of high-level optimization systems (wrappers) for relevance determination and constructive induction, and on integrating these wrappers with elicited knowledge on attribute relevance and synthesis. In particular, we discuss decision support issues for the application (cost prediction for automobile insurance markets in several states) and report experiments using i>D2K, a Java-based visual programming system for data mining and information visualization, and several commercial and research tools. We describe exploratory clustering, descriptive statistics, and supervised decision tree learning in this application, focusing on a parallel genetic algorithm (GA) system, i>Jenesis, which is used to implement relevance determination (attribute subset selection). Deployed on several high-performance network-of-workstation systems (Beowulf clusters), i>Jenesis achieves a linear speedup, due to a high degree of task parallelism. Its test set accuracy is significantly higher than that of decision tree inducers alone and is comparable to that of the best extant search-space based wrappers.

Inductive learning techniques in design process -- A design concept learning system

This paper discusses the role of inductive learning in the context of intelligent design support. It argues that inductive learning is an integral part of conceptual design process and that computer-based design systems can be enhanced by the integra

Inductive learning techniques in design process -- A design concept learning system

This paper discusses the role of inductive learning in the context of intelligent design support. It argues that inductive learning is an integral part of conceptual design process and that compute...

Inductive learning of decision trees: application to fault isolation of an induction motor

This work deals with fault detection and isolation (FDI) of an induction motor. Its supervision cannot be performed on the sole knowledge of analytical redundancy relations : a normal functioning state of the motor and a speed-sensor failure state cannot be distinguished from a behavioral analytical model. A solution is proposed using two inductive learning techniques based on decision-tree formalism: C4.5 which is a milestone in top–down induction of decision trees, and BUST which is a solution for the functional separability problem of decision trees.

A knowledge-based system for patient image pre-fetching in heterogeneous database environments - modeling, design, and evaluation

When performing primary reading on a newly taken radiological examination, a radiologist often needs to reference relevant prior images of the same patient for confirmation or comparison purposes. Support of such image references is of clinical importance and may have significant effects on radiologists' examination reading efficiency, service quality, and work satisfaction. To effectively support such image reference needs, we proposed and developed a knowledge-based patient image pre-fetching system, addressing several challenging requirements of the application that include representation and learning of image reference heuristics and management of data-intensive knowledge inferencing. Moreover, the system demands an extensible and maintainable architecture design capable of effectively adapting to a dynamic environment characterized by heterogeneous and autonomous data source systems. In this paper, we developed a synthesized object-oriented entity- relationship model, a conceptual model appropriate for representing radiologists' prior image reference heuristics that are heuristic oriented and data intensive. We detailed the system architecture and design of the knowledge-based patient image pre-fetching system. Our architecture design is based on a client-mediator-server framework, capable of coping with a dynamic environment characterized by distributed, heterogeneous, and highly autonomous data source systems. To adapt to changes in radiologists' patient prior image reference heuristics, ID3-based multidecision-tree induction and CN2-based multidecision induction learning techniques were developed and evaluated. Experimentally, we examined effects of the pre-fetching system we created on radiologists' examination readings. Preliminary results show that the knowledge-based patient image pre-fetching system more accurately supports radiologists' patient prior image reference needs than the current practice adopted at the study site and that radiologists may become more efficient, consultatively effective, and better satisfied when supported by the pre-fetching system than when relying on the study site's pre-fetching practice.

Automating the analysis of option pricing algorithms through intelligent knowledge acquisition approaches

The traditional approach for estimating the performance of numerical methods is to combine an operation's count with an asymptotic error analysis. This analytic approach gives a general feel of the comparative efficiency of methods, but it rarely leads to very precise results. It is now recognized that accurate performance evaluation can be made only with actual measurements on working software. Given that such an approach requires an enormous amount of performance data related to actual measurements, the development of novel approaches and systems that intelligently and efficiently analyze these data is of great importance to scientists and engineers. The paper presents intelligent knowledge acquisition approaches and an integrated prototype system, which enables the automatic and systematic analysis of performance data. The system analyzes the performance data which is usually stored in a database with statistical, and inductive learning techniques and generates knowledge which can be incorporated in a knowledge base incrementally. We demonstrate the use of the system in the context of a case study, covering the analysis of numerical algorithms for the pricing of American vanilla options in a Black and Scholes modeling framework. We also present a qualitative and quantitative comparison of two techniques used for the automated knowledge acquisition phase.

Cloning biological synergies improves control of elbow neuroprosthesis.

The control method proposed in this article is based on the following hypotheses: once the task and preferred strategy for movement are selected, then the voluntary (natural) control drives the proximal segment (shoulder joint), and the synergistic (artificial) control drives the distal segment (elbow joint). This concept relies on a simulation of the natural control of the extremities. The literature suggests, and the authors confirmed in experiments, that reproducible synergies between shoulder and elbow joint movement exist. The authors describe a method that uses an applied inductive learning (IL) technique for determining synergies between movement of joints while reaching. This method relies on an hierarchical mutual information classifier algorithm. The synergy is shown in a map of flexion/extension (F/E) angular velocities at the shoulder and elbow joints. Two other shoulder joint rotations (abduction/adduction and humeral rotation) are independent of the F/E synergy, thus indicating that the results of this study are applicable to three-dimensional (3-D) movement.

Concept development of consumer goods utilizing strategic knowledge

This paper proposes that knowledge systems should have the layered structure which consists of 'strategy', 'tactics', and 'physical world'. Based on this structure, we propose a novel method for the computer aided strategic concept formation in the early conceptual stages of a product1 development process. This method enables the construction of computer-aided representations of a product concept along with strategy and tactics, while it stimulates a person to articulate his/her idea in mind. The system called BICSS has been developed based on this method. The basic ideas of the system are: (1) to treat the strategic knowledge and tactical knowledge independently; and (2) to repeat the interactive cycle between humans and computers. A simple expert system is used for strategic knowledge. The interactive evolutionary computing (IEC) including interactive Genetic Algorithms and inductive learning techniques is used for tactical knowledge. The proposed method has been qualitatively validated by a case study.

Auto-bias selection for developing learning-based scheduling systems

Inductive learning techniques have shown encouraging achievements in reducing the effort for knowledge acquisition in the development of knowledge-based scheduling systems. However, there is little research on selecting the proper learning biases to facilitate the development of knowledge bases (KBs), so that the prediction accuracy of the resulting KBs is enhanced. In this study, we propose an auto-bias selection procedure that is capable of determining good learning biases, such as a proper feature set (system attributes) and a suitable learning algorithm. Through the case study conducted, the proposed procedure shows its capability to select proper learning biases and the results of experiments demonstrate that the scheduling KBs, developed by the learning biases selected, have a superior ability to produce the correct scheduling strategy over the scheduling KBs developed without exploring the proper learning biases.

Three machine learning techniques for automatic determination of rules to control locomotion.

Automatic prediction of gait events (e.g., heel contact, flat foot, initiation of the swing, etc.) and corresponding profiles of the activations of muscles is important for real-time control of locomotion. This paper presents three supervised machine learning (ML) techniques for prediction of the activation patterns of muscles and sensory data, based on the history of sensory data, for walking assisted by a functional electrical stimulation (FES). Those ML's are: 1) a multilayer perceptron with Levenberg-Marquardt modification of backpropagation learning algorithm; 2) an adaptive-network-based fuzzy inference system (ANFIS); and 3) a combination of an entropy minimization type of inductive learning (IL) technique and a radial basis function (RBF) type of artificial neural network with orthogonal least squares learning algorithm. Here we show the prediction of the activation of the knee flexor muscles and the knee joint angle for seven consecutive strides based on the history of the knee joint angle and the ground reaction forces. The data used for training and testing of ML's was obtained from a simulation of walking assisted with an FES system [39]. The ability of generating rules for an FES controller was selected as the most important criterion when comparing the ML's. Other criteria such as generalization of results, computational complexity, and learning rate were also considered. The minimal number of rules and the most explicit and comprehensible rules were obtained by ANFIS. The best generalization was obtained by the IL and RBF network.

Conceptual clustering on partitioned data: Tree-Weaver

Most knowledge discovery in databases (KDD) research is concentrated on supervised inductive learning. Conceptual clustering is an unsupervised inductive learning technique that organizes observations into an abstraction hierarchy without using predefined class values. However, a typical conceptual clustering algorithm is not suitable for a KDD task because of space and time constraints. Furthermore, typical incremental and non-incremental clustering algorithms are not designed for a partitioned data set. In this paper, we present a conceptual clustering algorithm that works on partitioned data. The proposed algorithm improves the clustering process by using less computation time and less space while maintaining the clustering quality.

A machine learning approach to inductive query by examples: An experiment using relevance feedback, ID3, genetic algorithms, and simulated annealing

Information retrieval using probabilistic techniques has attracted significant attention on the part of researchers in information and computer science over the past few decades. In the 1980s, knowledge-based techniques also made an impressive contribution to “intelligent” information retrieval and indexing. More recently, information science researchers have turned to other newer inductive learning techniques including symbolic learning, genetic algorithms, and simulated annealing. These newer techniques, which are grounded in diverse paradigms, have provided great opportunities for researchers to enhance the information processing and retrieval capabilities of current information systems. In this article, we first provide an overview of these newer techniques and their use in information retrieval research. In order to familiarize readers with the techniques, we present three promising methods: The symbolic ID3 algorithm, evolution-based genetic algorithms, and simulated annealing. We discuss their knowledge representations and algorithms in the unique context of information retrieval. An experiment using a 8000-record COMPEN database was performed to examine the performances of these inductive query-by-example techniques in comparison with the performance of the conventional relevance feedback method. The machine learning techniques were shown to be able to help identify new documents which are similar to documents initially suggested by users, and documents which contain similar concepts to each other. Genetic algorithms, in particular, were found to out-perform relevance feedback in both document recall and precision. We believe these inductive machine learning techniques hold promise for the ability to analyze users' preferred documents (or records), identify users' underlying information needs, and also suggest alternatives for search for database management systems and Internet applications. © 1998 John Wiley & Sons, Inc.