Vladimir Kurbalija

The influence of global constraints on similarity measures for time-series databases

A time series consists of a series of values or events obtained over repeated measurements in time. Analysis of time series represents an important tool in many application areas, such as stock-market analysis, process and quality control, observation of natural phenomena, and medical diagnosis. A vital component in many types of time-series analyses is the choice of an appropriate distance/similarity measure. Numerous measures have been proposed to date, with the most successful ones based on dynamic programming. Being of quadratic time complexity, however, global constraints are often employed to limit the search space in the matrix during the dynamic programming procedure, in order to speed up computation. Furthermore, it has been reported that such constrained measures can also achieve better accuracy. In this paper, we investigate four representative time-series distance/similarity measures based on dynamic programming, namely Dynamic Time Warping (DTW), Longest Common Subsequence (LCS), Edit distance with Real Penalty (ERP) and Edit Distance on Real sequence (EDR), and the effects of global constraints on them when applied via the Sakoe-Chiba band. To better understand the influence of global constraints and provide deeper insight into their advantages and limitations we explore the change of the 1-nearest neighbor graph with respect to the change of the constraint size. Also, we examine how these changes reflect on the classes of the nearest neighbors of time series, and evaluate the performance of the 1-nearest neighbor classifier with respect to different distance measures and constraints. Since we determine that constraints introduce qualitative differences in all considered measures, and that different measures are affected by constraints in various ways, we expect our results to aid researchers and practitioners in selecting and tuning appropriate time-series similarity measures for their respective tasks.

A framework for time-series analysis

The popularity of time-series databases in many applications has created an increasing demand for performing data-mining tasks (classification, clustering, outlier detection, etc.) on time-series data. Currently, however, no single system or library exists that specializes on providing efficient implementations of data-mining techniques for time-series data, supports the necessary concepts of representations, similarity measures and preprocessing tasks, and is at the same time freely available. For these reasons we have designed a multi-purpose, multifunctional, extendable system FAP - Framework for Analysis and Prediction, which supports the aforementioned concepts and techniques for mining time-series data. This paper describes the architecture of FAP and the current version of its Java implementation which focuses on time-series similarity measures and nearest-neighbor classification. The correctness of the implementation is verified through a battery of experiments which involve diverse time-series data sets from the UCR repository.

The Influence of Global Constraints on DTW and LCS Similarity Measures for Time-Series Databases

Analysis of time series represents an important tool in many application areas. A vital component in many types of time-series analysis is the choice of an appropriate distance/similarity measure. Numerous measures have been proposed to date, with the most successful ones based on dynamic programming. Being of quadratic time complexity, however, global constraints are often employed to limit the search space in the matrix during the dynamic programming procedure, in order to speed up computation. In this paper, we investigate two representative time-series distance/similarity measures based on dynamic programming, Dynamic Time Warping (DTW) and Longest Common Subsequence (LCS), and the effects of global constraints on them. Through extensive experiments on a large number of time-series data sets, we demonstrate how global constrains can significantly reduce the computation time of DTW and LCS. We also show that, if the constraint parameter is tight enough (less than 10–15% of time-series length), the constrained measure becomes significantly different from its unconstrained counterpart, in the sense of producing qualitatively different 1-nearest neighbour (1NN) graphs. This observation highlights the need for careful tuning of constraint parameters in order to achieve a good trade-off between speed and accuracy.

Comparison of different weighting schemes for the kNN classifier on time-series data

Many well-known machine learning algorithms have been applied to the task of time-series classification, including decision trees, neural networks, support vector machines and others. However, it was shown that the simple 1-nearest neighbor (1NN) classifier, coupled with an elastic distance measure like Dynamic Time Warping (DTW), often produces better results than more complex classifiers on time-series data, including k-nearest neighbor (kNN) for values of

Matching Observed with Empirical Reality --What you see is what you get?

k>1

Multiple Sclerosis Diagnoses--Case-Base Reasoning Approach

k>1. In this article, we revisit the kNN classifier on time-series data by considering ten classic distance-based vote weighting schemes in the context of Euclidean distance, as well as four commonly used elastic distance measures: DTW, Longest Common Subsequence, Edit Distance with Real Penalty and Edit Distance on Real sequence. Through experiments on the complete collection of UCR time-series datasets, we confirm the view that the 1NN classifier is very hard to beat. Overall, for all considered distance measures, we found that variants of the Dudani weighting scheme produced the best results.

An Application of Case-Based Reasoning in Multidimensional Database Architecture

There exists a major concern regarding toxic effects of immunosuppressive medication on the kidney graft during post-transplant care, with observed variation in individual susceptibility to adverse drug effects amongst patients. To date, there has been no possibility to identify susceptible patients prospectively. This study analyzes medical data which includes time series of measures of renal function and trough levels of immunosuppressive drug Tacrolimus, with the main aim of identifying patients susceptible to drug toxicity. We evaluate a plethora of time-series distance measures, determining their appropriateness to the domain based on two criteria: (1) preserving the expected correlations between distances, and (2) ability to detect the expected patterns of interaction between immunosuppressive drug levels and renal function. Besides identifying the most suitable time-series distance measures, we observed that the majority of patients do not exhibit an association between impaired graft function and higher Tacrolimus dosing. On the other hand, the minority of patients determined most sensitive to varying Tacrolimus levels showed a strong tendency to prefer low Tacrolimus dosing.

Computer Science and Information Systems: Publishing an International Open Access Journal in a Developing Country

This paper outlines the primary steps to investigate if artificial agents can be considered as true substitutes of humans. Based on a Socially augmented microworld SAM human tracking behavior was analyzed using time series. SAM involves a team of navigators jointly steering a driving object along different virtual tracks containing obstacles and forks. Speed and deviances from track are logged, producing high-resolution time series of individual training and cooperative tracking behavior. In the current study 52 time series of individual tracking behavior on training tracks were clustered according to different similarity measures. Resulting clusters were used to predict cooperative tracking behavior in fork situations. Results showed that prediction was well for tracking behavior shown at the first and, moderately well at the third fork of the cooperative track: navigators switched from their trained to a different tracking style and then back to their trained behavior. This matches with earlier identified navigator types, which were identified on visual examination. Our findings on navigator types will serve as a basis for the development of artificial agents, which can be compared later to behavior of human navigators.