Analysis of Similarity Measures in Times Series Clustering for the Discovery of Building Energy Patterns

Abstract

Forecasting and modeling building energy profiles require tools able to discover patterns within large amounts of collected information. Clustering is the main technique used to partition data into groups based on internal and a priori unknown schemes inherent of the data. The adjustment and parameterization of the whole clustering task is complex and submitted to several uncertainties, being the similarity metric one of the first decisions to be made in order to establish how the distance between two independent vectors must be measured. The present paper checks the effect of similarity measures in the application of clustering for discovering representatives in cases where correlation is supposed to be an important factor to consider, e.g., time series. This is a necessary step for the optimized design and development of efficient clustering-based models, predictors and controllers of time-dependent processes, e.g., building energy consumption patterns. In addition, clustered-vector balance is proposed as a validation technique to compare clustering performances.

Electricity Customer Clustering Following Experts’ Principle for Demand Response Applications

“…In their work, energy consumption graphs were interpreted as a mixture of Gaussian distributions, and the distance between load profiles was defined by the K-L divergence of the Gaussian mixture distributions. In the literature, other various machine learning techniques such as the self-organizing map (SOM), neural network, support vector clustering, dynamic time warping (DTW), and latent Dirichlet allocation (LDA) have also been applied to cluster electricity customers [9][10][11][12][13][14][15][16]. Furthermore, the transformation of input data to other domains such as frequency domain has been proposed in the literature [17].…”

Vehicle Recognition in Aerial Lidar Point Cloud Based on Dynamic Time Warping

“…The minimized warping cost is considered as the similarity measurement (Iglesias et al 2013, Keogh et al 2005). …”

“…The modification provides an opportunity to detect vehicles that are entirely or partially covered by trees. In the second step, the vehicle class is recognized by the similarity between the profile shape of the potential vehicle segment and a directed real vehicle profile based on dynamic time warping (Iglesias et al 2013, Keogh et al 2005. Other aspects, such as location and size, can also be estimated from the vehicle segment.…”

Clustering of OECD Countries Out of Pocket Health Expenditure Time Series Data

“…For time series data comparison where trends and evolutions are intended to be evaluated, or when the shape formed by the ordered succession of features is relevant, similarity measures based on Pearson's correlation have also been utilized (Iglesias & Kastner, 2013). The dissimilarity forms of correlation coefficients are represented below (Glynn, 2005): Dissimilarity = 1-Correlation, Dissimilarity = (1-Correlation)/2, Dissimilarity = 1-Abs (Correlation), Dissimilarity = Sqrt (1-Correlation )…”