Intelligent Query Answering with Virtual Mining and Materialized Views

Querying a database is a common task for traditional database systems. Producing answers effectively depends largely on users' knowledge about the query language and the database schema. In order to improve effectiveness and convenience of querying databases, we design a multi-agent system working cooperatively in an intelligent way to analyze user's request and revise the query with virtual mining and materialized views. Virtual mining views are data mining rules discovered from the database and materialized views are pre-computed data. This paper presents work in progress on the implementation and preliminary efficiency tests of the proposed system. The experimental results demonstrate the effectiveness of our multi-agent system in answering queries sharing the same pattern.

[1]  Chen Li,et al.  Generating efficient plans for queries using views , 2001, SIGMOD '01.

[2]  Rada Chirkova,et al.  Query evaluation using overlapping views: completeness and efficiency , 2006, SIGMOD Conference.

[3]  Alon Y. Halevy,et al.  Answering queries using views: A survey , 2001, The VLDB Journal.

[4]  Ion Muslea,et al.  Machine learning for online query relaxation , 2004, KDD.

[5]  Surajit Chaudhuri.  Generalization and a framework for query modification , 1990, [1990] Proceedings. Sixth International Conference on Data Engineering.

[6]  Toon Calders,et al.  Mining Views: Database Views for Data Mining , 2008, 2008 IEEE 24th International Conference on Data Engineering.

[7]  Jiawei Han,et al.  Intelligent Query Answering by Knowledge Discovery Techniques , 1996, IEEE Trans. Knowl. Data Eng..

[8]  Toon Calders,et al.  Integrating Pattern Mining in Relational Databases , 2006, PKDD.

[9]  Alvaro A. A. Fernandes,et al.  Logic-Based Integration of Query Answering and Knowledge Discovery , 2004, FQAS.

[10]  Qiming Chen,et al.  A Structured Approach for Cooperative Query Answering , 1994, IEEE Trans. Knowl. Data Eng..

[11]  Tsau Young Lin,et al.  Intelligent query answering based on neighborhood systems and data mining techniques , 2004, Proceedings. International Database Engineering and Applications Symposium, 2004. IDEAS '04..