People

Publications

Datasets and Data Generators

Supported by National Science Foundation

Award Number: IIS-0133825

PI: George Kollios

The goal of this project was to develop new indexing and data mining methods for spatio-temporal databases and related applications. In particular, we concentrated on the following problems:

(a) Efficient Indexing of Spatio-temporal Datasets: There are at least two kinds of interesting queries in such an environment, namely ``Future'' and ``Historical'' queries. If the functions by which object move/change are known, we can answer queries about the objects' anticipated positions/extent in the future. The answer to such queries is based on the time the query is executed. On the other hand, if the past locations of moving objects are stored in a database, we are interested in providing efficient index structures for querying the past. We developed methods to answer efficiently both types of queries. 

(b) Mining Spatio-temporal Databases: Spatio-temporal data are usually noisy and complex. We investigated methods and algorithms for efficient computation of appropriate similarity models and data mining operations (like similarity indexing) in this environment. Also, we have developed efficient and effective methods to store and query spatio-temporal data warehouses for a number of different aggregation queries and finding approximate periodic patterns in large trajectory datasets. In addition, we have created synthetic datasets and generators for benchmarking and testing the proposed algorithms.

(c) Recently, we have been investigating robust and effective methods to collect spatio-temporal data from a collection of sensors distributed in the environment and process and summarize spatio-temporal data streams. In addition, we considered the problem of authentication and verification of streaming and non-streaming data both in spatio-temporal and relational environments.