Suchergebnisse

Time series are ubiquitous, and a measure to assess their similarity is a core part of many computational systems. In particular, the similarity measure is the most essential ingredient of time series clustering and classification systems. Because of this importance, countless approaches to estimate time series similarity have been proposed. However, there is a lack of comparative studies using empirical, rigorous, quantitative, and large-scale assessment strategies. In this article, we provide an extensive evaluation of similarity measures for time series classification following the aforementioned principles. We consider 7 different measures coming from alternative measure 'families', and 45 publicly-available time series data sets coming from a wide variety of scientific domains. We focus on out-of-sample classification accuracy, but in-sample accuracies and parameter choices are also discussed. Our work is based on rigorous evaluation methodologies and includes the use of powerful statistical significance tests to derive meaningful conclusions. The obtained results show the equivalence, in terms of accuracy, of a number of measures, but with one single candidate outperforming the rest. Such findings, together with the followed methodology, invite researchers on the field to adopt a more consistent evaluation criteria and a more informed decision regarding the baseline measures to which new developments should be compared. © 2014 Elsevier B.V. All rights reserved. ; We thank the people who made available or contributed to the UCR time series repository. This research has been funded by 2009-SGR-1434 from Generalitat de Catalunya, JAEDOC069/2010 from Consejo Superior de Investigaciones Científicas, and TIN2009-13692-C03-01 and TIN2012-38450-C03-03 from the Spanish Government, and EU Feder funds. Funding Details ; Peer Reviewed

The cognitive prognosis of acquired brain injury (ABI) patients is a valuable tool for an improved and personalized treatment. In this paper, we explore the task of automatic cognitive prognosis of ABI patients via machine learning techniques. Based on a set of pre-treatment assessments, distinct classifiers are trained to predict whether the patient will improve in one or any of three cognitive areas: attention, memory, and executive functioning. Results show that variables such as the age at the moment of the injury, the patient's etiology, or the neuropsychological evaluation scores obtained before the treatment are relevant for prognosis and easily yield statistically significant accuracies. Additionally, the prognostic relevance of these and other variables is studied by means of standard feature selection methodologies. The outputs of the present paper add to the discussion on current cognitive rehabilitation practices and push towards the exploitation of existing technologies for improving medical evaluations and treatments. ; We thank all the patients and staff from Institut Guttmann who cooperated in data collection. This work has been partially funded by TIN-2012-38450-C03-03 from the Spanish Government (all authors), JAEDOC069/2010 from Consejo Superior de Investigaciones Cientıficas (J.S.), and 2009-SGR-1434 from Generalitat de Catalunya ; Peer Reviewed