Skip to Main Content
Time series data are often found in diverse fields, such as science, business, medicine and engineering. In this paper, we focus on sequential pattern mining for categorical time series datasets that contain multiple independent timeseries. Frequent patterns are considered important in many applications. However, collected data are generally afflicted with noise. Conventional sequential pattern mining methods that use exact matching may meet difficulties in mining databases with long sequences and noise. We propose a framework that uncovers frequent approximate sequential patterns with multiple widths. A mined pattern in this framework is a representative of a group of sequences (with various widths) that follow the pattern's event flow order. The presentation of the patterns also gives insight into the occurrence of the pattern longitudinally and across the population. The pattern can be recognized as a common pattern across the multiple time series, time, or both. We name this novel framework MWASP: Multiple-Width Approximate Sequential Patterns.