Unsupervised pattern mining from symbolic temporal data

摘要: We present a unifying view of temporal concepts and data models in order to categorize existing approaches for unsupervised pattern mining from symbolic data. In particular we distinguish time point-based methods interval-based as well univariate multivariate methods. The paradigms the robustness many proposed are compared aid selection appropriate method given problem. For points, sequential algorithms can be used express equality points with gaps limited suffix tree more efficient. Recently, efficient have been mine general concept partial points. interval precise start end relations Allen formulate patterns. recently Time Series Knowledge Representation is robust on noisy offers an alternative semantic that avoids ambiguity expressive. both languages proposed.