Fast mining of massive tabular data via approximate distance computations
UNSPECIFIED (2002) Fast mining of massive tabular data via approximate distance computations. In: 18th International Conference on Data Engineering, FEB 26-MAR 01, 2002, SAN JOSE, CA.Full text not available from this repository.
Tabular data abound in many data stores: traditional relational databases store tables, and new applications also generate massive tabular datasets. For example, consider the geographic distribution of cell phone traffic at different base stations across the country or the evolution of traffic at Internet routers over time. Detecting similarity patterns in such data sets (e.g., which geographic regions have similar cell phone usage distribution, which IP subnet traffic distributions over tithe intervals are similar, etc) is of great importance. Identification of such patterns poses many conceptual challenges (what is a suitable similarity distance function for two "regions") as well as technical challenges (how to perform similarity computations efficiently as massive tables get accumulated over time) that we address. We present methods for determining similar regions in massive tabular data. Our methods are for computing the "distance" between any two subregions of a tabular data: they are approximate, but highly accurate as ye prove mathematically, and they are fast, running in tithe nearly linear in the table size. Our methods are general since these distance computations can be applied to arty raining or similarity algorithms that use L-p norms. A novelty of our distance computation procedures is that they work for an v L-p norms-not only the traditionally p = 2 or p = 1, but for all p less than or equal to 2; the choice of p, say fractional p, provides an interesting alternative similarity, behavior! We rise our algorithms in a detailed experimental study of the clustering patterns in real tabular data obtained front one of AT&T's data stores and show that our methods are substantially faster than straightforward methods while remaining highly accurate, and able to detect interesting patterns by varying the value of p.
|Item Type:||Conference Item (UNSPECIFIED)|
|Subjects:||Q Science > QA Mathematics > QA76 Electronic computers. Computer science. Computer software
T Technology > TK Electrical engineering. Electronics Nuclear engineering
|Series Name:||PROCEEDINGS OF THE INTERNATIONAL CONFERENCE ON DATA ENGINEERING (SERIES)|
|Journal or Publication Title:||18TH INTERNATIONAL CONFERENCE ON DATA ENGINEERING, PROCEEDINGS|
|Publisher:||IEEE COMPUTER SOC|
|Editor:||Agrawal, R and Dittrich, K and Ngu, AHH|
|Number of Pages:||10|
|Page Range:||pp. 605-614|
|Title of Event:||18th International Conference on Data Engineering|
|Location of Event:||SAN JOSE, CA|
|Date(s) of Event:||FEB 26-MAR 01, 2002|
Actions (login required)