Analysis of Time Series Data with Predictive Clustering Trees

Abstract

Abstract. Predictive clustering is a general framework that unifies clustering and prediction. This paper investigates how to apply this framework to cluster time series data. The resulting system, Clus-TS, constructs predictive clustering trees (PCTs) that partition a given set of time series into homogeneous clusters. In addition, PCTs provide a symbolic description of the clusters. We evaluate Clus-TS on time series data from microarray experiments. Each data set records the change over time in the expression level of yeast genes in response to a change in environmental conditions. Our evaluation shows that Clus-TS is able to cluster genes with similar responses, and to predict the time series based on the description of a gene. Clus-TS is part of a larger project where the goal is to investigate how global models can be combined with inductive databases.

Hierarchical annotation of medical images

“…With instantiation of the variance and prototype function the PCTs can handle different types of data, e.g. multiple targets [8] or time series [9]. A detailed description of the PCT framework can be found in [7].…”

Human Body Shape Prediction and Analysis Using Predictive Clustering Tree

“…A Predictive Clustering Tree (PCT) introduced by Dzeroski et al [12] allows for predicting multiple target variables. Human body shape in a statistical shape space can be represented by multiple variables like PCA coordinates.…”

“…At each node, the PCT tries to find the best partition rule using D through minimizing the variations of clustered T . As a result, each leaf node represents a cluster, and the conjunction of the conditions on the path from the root to the leaf forms the induction rule [12].…”

“…It calls on Algorithm 2 to find the best partition for the current unclustered training set [12]. Using the returned partitioning condition P test, we create a new node and a local partition P arts.…”

Motif-Based Classification of Time Series with Bayesian Networks and SVMs

“…Motif-based Classification. Motifs have been used for sequence classification in biological domain [6,15,8]. This is usually done in two steps: i) first motifs are extracted, then ii) each time series is represented as an attribute vector using motifs so that a classifier like SVM [15], Naive Bayes [8], Decision Tree [6], etc.…”