An empirical evaluation of similarity measures for time series classification

Serrí, Joan; Arcos, Josep Lluís

doi:10.1016/j.knosys.2014.04.035

Cited by 199 publications

(128 citation statements)

References 39 publications

Supporting

Mentioning

128

Contrasting

Order By: Relevance

“…In [20,21], principal component analysis and similarity measures for time series evaluation of generic data are discussed. The conclusions are applicable to smart meter data, although the method works best with fewer meters than recordings and thus, conversely, the dataset expands.…”

Section: Literature Reviewmentioning

confidence: 99%

Electricity Consumption Clustering Using Smart Meter Data

2018

View full text Add to dashboard Cite

Electricity smart meter consumption data is enabling utilities to analyze consumption information at unprecedented granularity. Much focus has been directed towards consumption clustering for diversifying tariffs; through modern clustering methods, cluster analyses have been performed. However, the clusters developed exhibit a large variation with resulting shadow clusters, making it impossible to truly identify the individual clusters. Using clearly defined dwelling types, this paper will present methods to improve clustering by harvesting inherent structure from the smart meter data. This paper clusters domestic electricity consumption using smart meter data from the Danish city of Esbjerg. Methods from time series analysis and wavelets are applied to enable the K-Means clustering method to account for autocorrelation in data and thereby improve the clustering performance. The results show the importance of data knowledge and we identify sub-clusters of consumption within the dwelling types and enable K-Means to produce satisfactory clustering by accounting for a temporal component. Furthermore our study shows that careful preprocessing of the data to account for intrinsic structure enables better clustering performance by the K-Means method.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

Electricity Consumption Clustering Using Smart Meter Data

2018

View full text Add to dashboard Cite

show abstract

“…Therefore, we do not intend to compare the performance of different classifiers and similarity metrics (which have been empirically studied in [26]). We rely in our study on the most frequently used classification method: Nearest Neighbor Classification (1NN) with DTW distance.…”

Section: Setup and Datasetsmentioning

confidence: 99%

Time Series Classification with Discrete Wavelet Transformed Data

Bissyandé

Klein

et al. 2016

Int. J. Soft. Eng. Knowl. Eng.

View full text Add to dashboard Cite

Time series mining has become essential for extracting knowledge from the abundant data that flows out from many application domains. To overcome storage and processing challenges in time series mining, compression techniques are being used. In this paper, we investigate the loss/gain of performance of time series classification approaches when fed with lossy-compressed data. This extended empirical study is essential for reassuring practitioners, but also for providing more insights on how compression techniques can even be effective in smoothing and reducing noise in time series data. From a knowledge engineering perspective, we show that time series may be compressed by 90% using discrete wavelet transforms and still achieve remarkable classification accuracy, and that residual details left by popular wavelet compression techniques can sometimes even help achieve higher classification accuracy than the raw time series data, as they better capture essential local features.

show abstract

“…Thanks to its maturity and performance, DTW based 1NN classification has become one of the most prominent TSC approaches. Although recent empirical comparison [23] reveals that the Time Warp Edit Distance (TWED) [15] performs statistically more accurate than DTW. Other common distance measures include Euclidean distance, Edit Distance on Real sequence (EDR) [5] and Minimum Jump Cost (MJC) [24].…”

Section: Related Workmentioning

confidence: 99%

DSCo: A Language Modeling Approach for Time Series Classification

Bissyandé

et al. 2016

Machine Learning and Data Mining in Pattern Recognition

View full text Add to dashboard Cite

Abstract. Time series data are abundant in various domains and are often characterized as large in size and high in dimensionality, leading to storage and processing challenges. Symbolic representation of time series -which transforms numeric time series data into texts -is a promising technique to address these challenges. However, these techniques are essentially lossy compression functions and information are partially lost during transformation. To that end, we bring up a novel approach named Domain Series Corpus (DSCo), which builds per-class language models from the symbolized texts. To classify unlabeled samples, we compute the fitness of each symbolized sample against all per-class models and choose the class represented by the model with the best fitness score. Our work innovatively takes advantage of mature techniques from both time series mining and NLP communities. Through extensive experiments on an open dataset archive, we demonstrate that it performs similarly to approaches working with original uncompressed numeric data.

show abstract

An empirical evaluation of similarity measures for time series classification

Cited by 199 publications

References 39 publications

Electricity Consumption Clustering Using Smart Meter Data

Electricity Consumption Clustering Using Smart Meter Data

Time Series Classification with Discrete Wavelet Transformed Data

DSCo: A Language Modeling Approach for Time Series Classification

Contact Info

Product

Resources

About