Relational Classification using Multiple View Approach with Voting

Abstract

is an important task in data mining and machine learning, in which a model is generated based on training dataset and that model is used to predict class label of unknown dataset. Various algorithms have been proposed to build accurate and scalable classifiers in data mining. These algorithms are only applied to single table. Today most real- world data are stored in relational format which is popular format for structured data which consist of tables connected via relations (primary key/ foreign key). So single table data mining algorithms cannot deal with relational data. To classify data from relational format need of multirelational classification arise which is used to analyze relational data and used to predict behaviour and unknown pattern automatically. For multirelational classification, various techniques are available which include upgrading existing algorithm, flatten relational data and multiple view approach. Multiple view approach learns from multiple views of a relational data and then combines the result of each view to classify unknown data. This paper presents proposed algorithm and experimental results for multiple view approach with voting as a view combination technique.