To date, the vast majority of research on time series data mining has focused on similarity search and to a lesser extent on clustering. However, from a knowledge discovery viewpoint, there are several important unsolved problems in time series data mining that are more interesting, important, and challenging. This project addresses these problems. The long-term goal is the creation of efficient algorithms to allow the extraction of knowledge in the form of patterns, anomalies, regularities and rules, from massive time series datasets.
Because of the ubiquity of times series data, the work may have benefits in areas as diverse as cardiology, industry, astronomy, medicine, bioinformatics, finance, meteorology, entertainment and networking. Local collaborators in industry and science, who will test the algorithms, have been identified. To enhance broader impacts of this project, results will be disseminated by an expansion of the UCR Time Series Data Mining Archive, which will make all code, datasets and publications created by the project freely available to data mining researchers and practitioners. The Web site (www.cs.ucr.edu/~eamonn/NSFcareer/NSF.html) provides more information about this project.
A special feature of the project is an effort involving undergraduate and graduate students to find, implement and compare relevant work to the approach developed in this project. Time series data analysis is very important for business and thus the project will have broad impact beyond its scientific impact.
