RESUMEN
Time Series Classification (TSC) involves building predictive models for a discrete target variable from ordered, real valued, attributes. Over recent years, a new set of TSC algorithms have been developed which have made significant improvement over the previous state of the art. The main focus has been on univariate TSC, i.e. the problem where each case has a single series and a class label. In reality, it is more common to encounter multivariate TSC (MTSC) problems where the time series for a single case has multiple dimensions. Despite this, much less consideration has been given to MTSC than the univariate case. The UCR archive has provided a valuable resource for univariate TSC, and the lack of a standard set of test problems may explain why there has been less focus on MTSC. The UEA archive of 30 MTSC problems released in 2018 has made comparison of algorithms easier. We review recently proposed bespoke MTSC algorithms based on deep learning, shapelets and bag of words approaches. If an algorithm cannot naturally handle multivariate data, the simplest approach to adapt a univariate classifier to MTSC is to ensemble it over the multivariate dimensions. We compare the bespoke algorithms to these dimension independent approaches on the 26 of the 30 MTSC archive problems where the data are all of equal length. We demonstrate that four classifiers are significantly more accurate than the benchmark dynamic time warping algorithm and that one of these recently proposed classifiers, ROCKET, achieves significant improvement on the archive datasets in at least an order of magnitude less time than the other three.
RESUMEN
Our hypothesis is that building ensembles of small sets of strong classifiers constructed with different learning algorithms is, on average, the best approach to classification for real-world problems. We propose a simple mechanism for building small heterogeneous ensembles based on exponentially weighting the probability estimates of the base classifiers with an estimate of the accuracy formed through cross-validation on the train data. We demonstrate through extensive experimentation that, given the same small set of base classifiers, this method has measurable benefits over commonly used alternative weighting, selection or meta-classifier approaches to heterogeneous ensembles. We also show how an ensemble of five well-known, fast classifiers can produce an ensemble that is not significantly worse than large homogeneous ensembles and tuned individual classifiers on datasets from the UCI archive. We provide evidence that the performance of the cross-validation accuracy weighted probabilistic ensemble (CAWPE) generalises to a completely separate set of datasets, the UCR time series classification archive, and we also demonstrate that our ensemble technique can significantly improve the state-of-the-art classifier for this problem domain. We investigate the performance in more detail, and find that the improvement is most marked in problems with smaller train sets. We perform a sensitivity analysis and an ablation study to demonstrate the robustness of the ensemble and the significant contribution of each design element of the classifier. We conclude that it is, on average, better to ensemble strong classifiers with a weighting scheme rather than perform extensive tuning and that CAWPE is a sensible starting point for combining classifiers.
RESUMEN
In the last 5 years there have been a large number of new time series classification algorithms proposed in the literature. These algorithms have been evaluated on subsets of the 47 data sets in the University of California, Riverside time series classification archive. The archive has recently been expanded to 85 data sets, over half of which have been donated by researchers at the University of East Anglia. Aspects of previous evaluations have made comparisons between algorithms difficult. For example, several different programming languages have been used, experiments involved a single train/test split and some used normalised data whilst others did not. The relaunch of the archive provides a timely opportunity to thoroughly evaluate algorithms on a larger number of datasets. We have implemented 18 recently proposed algorithms in a common Java framework and compared them against two standard benchmark classifiers (and each other) by performing 100 resampling experiments on each of the 85 datasets. We use these results to test several hypotheses relating to whether the algorithms are significantly more accurate than the benchmarks and each other. Our results indicate that only nine of these algorithms are significantly more accurate than both benchmarks and that one classifier, the collective of transformation ensembles, is significantly more accurate than all of the others. All of our experiments and results are reproducible: we release all of our code, results and experimental details and we hope these experiments form the basis for more robust testing of new algorithms in the future.
RESUMEN
This research is part of a wider project to build predictive models of bone age using hand radiograph images. We examine ways of finding the outline of a hand from an X-ray as the first stage in segmenting the image into constituent bones. We assess a variety of algorithms including contouring, which has not previously been used in this context. We introduce a novel ensemble algorithm for combining outlines using two voting schemes, a likelihood ratio test and dynamic time warping (DTW). Our goal is to minimize the human intervention required, hence we investigate alternative ways of training a classifier to determine whether an outline is in fact correct or not. We evaluate outlining and classification on a set of 1370 images. We conclude that ensembling with DTW improves performance of all outlining algorithms, that the contouring algorithm used with the DTW ensemble performs the best of those assessed, and that the most effective classifier of hand outlines assessed is a random forest applied to outlines transformed into principal components.
Asunto(s)
Determinación de la Edad por el Esqueleto/clasificación , Determinación de la Edad por el Esqueleto/métodos , Huesos de la Mano/diagnóstico por imagen , Mano/diagnóstico por imagen , Procesamiento de Imagen Asistido por Computador/métodos , Envejecimiento/fisiología , Algoritmos , Inteligencia Artificial , Automatización , Niño , Femenino , Dedos/anatomía & histología , Dedos/diagnóstico por imagen , Análisis de Fourier , Humanos , Funciones de Verosimilitud , Masculino , Análisis de Componente Principal , Estándares de Referencia , Programas Informáticos , Rayos XRESUMEN
BACKGROUND: The mechanics of small bowel motility are extremely complex. Routine clinical access to small bowel has been restricted to radiological enteric contrast studies which have not contributed significantly to the understanding of small bowel physiology. Small bowel mechanics are understood within a framework of individual visible or measurable elements such as peristaltic wave formation, intra-luminal pressure gradients and transit times. There are no global measures of small bowel function that can be readily obtained in vivo in humans. Magnetic resonance imaging (MRI) is playing an increasingly important role in radiological diagnosis of small bowel disease and dynamic MRI offers the possibility of capturing small bowel movement in three-dimensional cinematic datasets. The metrics that are used to describe small bowel mechanics, typically anatomical measures in isolated segments, are not suited to analysing these large dynamic datasets. The proposal in this paper is to leave behind all previously described anatomical metrics and to describe anew the mechanics of small bowel movement in mathematical terms derived from changes in pixel intensity within dynamic MRI datasets so that global small bowel activity might be summarised in a single novel metric. HYPOTHESIS: The hypothesis of this paper is that global small bowel activity can be quantified by a new dynamic MR based metric. EVALUATION: A proposed strategy for evaluation includes a progression through feasibility, optimisation, reliability and validation studies. Thereafter normal volunteers would be required in order to define normal ranges for the new metric. These ranges would describe small bowel activity during fasting or after ingestion of fluids and standard meals. Mathematical modelling of the data could follow a two stage approach. The first stage could be to study segmentation or extraction techniques by which the small bowel activity could be isolated from MRI signal generated by the rest of the abdomen. The second stage would be to apply a number of data mining techniques that would identify significant features within the datasets. CONCLUSION: If this approach proves to be a useful model for studying small bowel physiology in humans, it would afford significant new avenues of research and treatment particularly in areas such as enteric drug delivery, the ageing gut, and nutrition.