Efficiently finding unusual shapes in large image databases
Data Mining and Knowledge Discovery
Stock time series visualization based on data point importance
Engineering Applications of Artificial Intelligence
A review on time series data mining
Engineering Applications of Artificial Intelligence
Time series visualization based on shape features
Knowledge-Based Systems
Pattern discovery in data streams under the time warping distance
The VLDB Journal — The International Journal on Very Large Data Bases
| Hi-index | 0.00 |
Since their introduction in the seventies by Gibbs and McIntyre, dot plots have proved to be a powerful and intuitive technique for visual sequence analysis and mining. Their main domain of application is the field of bioinformatics where they are frequently used by researchers in order to elucidate genomic sequence similarities and alignment. However, this useful technique has remained comparatively constrained to domains where the data has an inherent discrete structure (i.e., text). In this paper we demonstrate how dot plots can be used for the analysis and mining of real-valued time series. We design a tool that creates highly descriptive dot plots which allow one to easily detect similarities, anomalies, reverse similarities, and periodicities as well as changes in the frequencies of repetitions. As the underlying algorithm scales well with the input size, we also show the feasibility of the plots for on-line data monitoring.