Frequency-Based Coverage Statistics Mining for Data Integration

Recent work in data integration has shown the importance of statistical information about the coverage and overlap of data sources for efficient query processing. Gathering and storing the required statistics presents many challenges, not the least of which is controlling the amount of statistics learned. In this paper we describe a novel application of data mining technology for statistics gathering. Specifically, we introduce a statistics mining approach which efficiently discovers frequently accessed query classes, and learns and stores statistics only with respect to these classes. We describe the details of our method, and present experimental results demonstrating the efficiency and effectiveness of our approach.

• Type: Conference
• Year: 2003
• Where: IJCAI
• Authors: Zaiqing Nie, Subbarao Kambhampati
• Keywords: Efficient Query Processing, IJCAI 2003, IJCAI 2007, Statistics, Statistics Mining Approach