Query Recommender System Using Hierarchical Classification

Abstract

In data warehouses, lots of data are gathered which are navigated and explored for analytical purposes. Even for expert people, to handle such a large data is a tough task. Handling such a voluminous data is more difficult task for non-expert users or for users who are not familiar with the database schema. The aim of this paper is to help this class of users by recommending them SQL queries that they might use. These SQL recommendations are selected by tracking the users past behavior and comparing them with other users. At first time, users may not know where to start their exploration. Secondly, users may overlook queries which help to retrieve important information. The queries are recorded and compared using hierarchical classification which is then re-ranked according to relevance. The relevant queries are retrieved using users querying behavior. Users use a query interface to issue a series of SQL queries that aim to analyze the data and mine it for interesting information.
DOI: 10.17762/ijritcc2321-8169.150676