Hierarchical rule generalisation for speaker identification in fiction books

This paper presents a hierarchical pattern matching and generalisation technique which is applied to the problem of locating the correct speaker of quoted speech found in fiction books. Patterns from a training set are generalised to create a small number of rules, which can be used to identify items of interest within the text. The pattern matching technique is applied to finding the Speech-Verb, Actor and Speaker of quotes found in ction books. The technique performs well over the training data, resulting in rule-sets many times smaller than the training set, but providing very high accuracy. While the rule-set generalised from one book is less effective when applied to different books than an approach based on hand coded heuristics, performance is comparable when testing on data closely related to the training set.

On 4 June 2008 Siberian Red Knots at Elbe Mouth kissed the canonical evening migration departure rule goodbye

Observations of departing Siberian-breeding Red Knots Calidris canutus canutus from their central staging site during northward migration, the Schleswig-Holstein Wadden Sea, Germany, in early June 2008, challenge the established notion that departing long-distance migrating waders only leave around sunset. During four days we scanned several thousand Red Knots for colour-ringed individuals and found a total of 20 different individuals that were previously ringed at either their main wintering site, the Banc d'Arguin in Mauritania, or at stopover sites on the Atlantic coast of France. Body masses of captured Red Knots in Schleswig-Holstein were higher than 200 g and hematocrite values showed an average of 58%, clearly indicating that they were ready for take-off. On all except one evening, we noted impressive departure movements during the incoming tide. On that exceptional evening a cold front thunderstorm passed over the area. Late the next morning, thousands of Red Knots departed during the incoming tide. We assume that the birds avoided taking off in adverse weather conditions and elaborate why Red Knots presumably traded off advantages from departing during twilight. We suggest that during spring migration, schedules are so tight that further delays decrease fitness, either because it would cause another full day of exposure to high predation risk by falcons, or because of conditions upon arrival on the tundra.

An approximate numerical technique for characterizing optical pulse propagation in inhomogeneous biological tissue

An approximate numerical technique for modeling optical pulse propagation through weakly scattering biological tissue is developed by solving the photon transport equation in biological tissue that includes varying refractive index and varying scattering/absorption coefficients. The proposed technique involves first tracing the ray paths defined by the refractive index profile of the medium by solving the eikonal equation using a Runge-Kutta integration algorithm. The photon transport equation is solved only along these ray paths, minimizing the overall computational burden of the resulting algorithm. The main advantage of the current algorithm is that it enables to discretise the pulse propagation space adaptively by taking optical depth into account. Therefore, computational efficiency can be increased without compromising the accuracy of the algorithm.

Review of approximate analyses of sheet forming processes

Approximate models are often used for the following purposes: in on-line control systems of metal forming processes where calculation speed is critical; to obtain quick, quantitative information on the magnitude of the main variables in the early stages of process design; to illustrate the role of the major variables in the process; as an initial check on numerical modelling; and as a basis for quick calculations on processes in teaching and training packages. The models often share many similarities; for example, an arbitrary geometric assumption of deformation giving a simplified strain distribution, simple material property descriptions - such as an elastic, perfectly plastic law - and mathematical short cuts such as a linear approximation of a polynomial expression. In many cases, the output differs significantly from experiment and performance or efficiency factors are developed by experience to tune the models. In recent years, analytical models have been widely used at Deakin University in the design of experiments and equipment and as a pre-cursor to more detailed numerical analyses. Examples that are reviewed in this paper include deformation of sandwich material having a weak, elastic core, load prediction in deep drawing, bending of strip (particularly of ageing steel where kinking may occur), process analysis of low-pressure hydroforming of tubing, analysis of the rejection rates in stamping, and the determination of constitutive models by an inverse method applied to bending tests.

A guided rule reduction system for prioritization of failures in fuzzy FMEA

Traditional Failure Mode and Effect Analysis (FMEA) utilizes the Risk Priority Number (RPN) ranking system to evaluate the risk level of failures, to rank failures, and to prioritize actions. Although this method is simple, it suffers from several shortcomings. In this paper, use of fuzzy inference techniques for RPN determination in an attempt to overcome the weaknesses associated with the traditional RPN ranking system is investigated. However, the fuzzy RPN model, suffers from the combinatorial rule explosion problem. As a result, a generic rule reduction approach, i.e. the Guided Rule Reduction System (GRRS), is proposed to reduce the number of rules that need to be provided by users during the fuzzy RPN modeling process. The proposed approach is evaluated using real-world case studies pertaining to semiconductor manufacturing. The results are analyzed, and implications of the proposed approach are discussed.

A modified fuzzy min-max neural network with a genetic-algorithm-based rule extractor for pattern classification

In this paper, a two-stage pattern classification and rule extraction system is proposed. The first stage consists of a modified fuzzy min-max (FMM) neural-network-based pattern classifier, while the second stage consists of a genetic-algorithm (GA)-based rule extractor. Fuzzy if-then rules are extracted from the modified FMM classifier, and a ??don't care?? approach is adopted by the GA rule extractor to minimize the number of features in the extracted rules. Five benchmark problems and a real medical diagnosis task are used to empirically evaluate the effectiveness of the proposed FMM-GA system. The results are analyzed and compared with other published results. In addition, the bootstrap hypothesis analysis is conducted to quantify the results of the medical diagnosis task statistically. The outcomes reveal the efficacy of FMM-GA in extracting a set of compact and yet easily comprehensible rules while maintaining a high classification performance for tackling pattern classification tasks.

On the use of fuzzy rule interpolation techniques for monotonic multi-input fuzzy rule base models

Constructing a monotonicity relating function is important, as many engineering problems revolve around a monotonicity relationship between input(s) and output(s). In this paper, we investigate the use of fuzzy rule interpolation techniques for monotonicity relating fuzzy inference system (FIS). A mathematical derivation on the conditions of an FIS to be monotone is provided. From the derivation, two conditions are necessary. The derivation suggests that the mapped consequence fuzzy set of an FIS to be of a monotonicity order. We further evaluate the use of fuzzy rule interpolation techniques in predicting a consequent associated with an observation according to the monotonicity order. There are several findings in this article. We point out the importance of an ordering criterion in rule selection for a multi-input FIS before the interpolation process; and hence, the practice of choosing the nearest rules may not be true in this case. To fulfill the monotonicity order, we argue with an example that conventional fuzzy rule interpolation techniques that predict each consequence separately is not suitable in this case. We further suggest another class of interpolation techniques that predicts the consequence of a set of observations simultaneously, instead of separately. This can be accomplished with the use of a search algorithm, such as the brute force, genetic algorithm or etc.

Fuzzy rule base assessment models : theoretical analyses and a case study

An assessment model is usually a mathematical model that produces a measuring index, in the form of a numerical score to a situation/object, with respect to the subject of measure. To allow a valid and useful comparison among various situations/objects according to their associated numerical scores to be made, two important properties, i.e., the monotone output property and output resolution properties, are essential in fuzzy inference-based assessment problems. In this paper, the conditions for a fuzzy assessment model to fulfill the monotone output property is investigated using a derivative approach. A guideline on how the input membership functions should be tuned is also provided. Besides, the output resolution property is defined as the derivative of the output of the assessment model with respect to the input, whereby the derivative should be greater than a minimum resolution. Based on the derivative, improvements to the output resolution property by refining the fuzzy production rules are suggested. A case study on the Bowles fuzzy RPN model to demonstrate the effectiveness of the properties is also included.