Approximate Query Answering with Frequent Sets and Maximum Entropy H. Mannila and P. Smyth   View Paper (PDF)   Return to OLAP, DW, and Data Mining Abstract We describe an approach to finding approximate answers to Boolean queries on 0/1 data. The basic idea is to build a probabilistic model for the data set and answer the queries on the basis of that model. The method consists of the following steps: 2. Model building: given a query, use the count information to construct an approximate probability model for the variables occurring in the query. DiSC'01 Copyright ©2002 ACM Inc.