Scalable Informative Rule Mining
Loading...
Date
2016-08-10
Authors
Feng, Guoyao
Advisor
Golab, Lukasz
Keshav, Srinivasan
Keshav, Srinivasan
Journal Title
Journal ISSN
Volume Title
Publisher
University of Waterloo
Abstract
In this thesis we present SIRUM: a system for Scalable Informative RUle Mining from multi-dimensional data. Informative rules have recently been studied in several contexts, including data summarization, data cube exploration and data quality. The objective is to produce a concise set of rules (patterns) over the values of the dimension attributes that provide the most information about the distribution of a numeric measure attribute. SIRUM optimizes this task for big, wide and distributed datasets. We implemented SIRUM in Spark and observed significant performance improvements on real data due to our optimizations.
Description
Keywords
Informative Rule Mining, Scalable Data Processing Systems