Mining generalized association rules and sequential patterns using SQL queries

Abstract

Database integration of mining is becoming in-creasingly important with tile installation of larger and larger data warehouses built around relational database technology. Most of the commercially available mining systems integrate loosely (typically, through an ODBC or SQL cursor interface) with data stored in DBMSs. In cases where the mining algorithm makes nmltiple passes over the data, it is also possible to cache the data in fiat files rather than retrieve mul-tiple times from the DBMS, to achieve better perfor-mance. Recent studies have found that for association rule mining, with carefully tuned SQL forinulations it is possible to achieve performance comparable to systems that cache the data in files outside the DBMS. The SQL implementation has potential for offering other quali-taUve advantages like automatic parallehzation, devel-opment ease, portability and inter-operability with re-lational operators. In"this paper, we present several alternatives for formulating as SQL queries association rule generalized to handle items with hierarchies on them and sequential pattern mining. This work illus-trates that it is possible to express computations that are significantly more complicated than simple boolean assom%tions, in SQL using essentially the same franm-work. 1