A new improved Approach for Feature Generation and Selection in multirelationalstatistical modelling using machine learning
Multi-relational classification is highly challengeable task in data mining, becauseso much data in our world is organised in multiple relations. The challenge comes from thehuge collection of search spaces and high calculation cost arises in the selection of featuredue to excessive complexity in the various relations. The state-of-the-art approach is based onclusters and inductive logical programming to retrieve important features and derivedhypothesis. However, those techniques are very slow and unable to create enough data andinformation to produce efficient classifiers. In the given paper, we proposed a fast andeffective method for the feature selection using multi-relational classification. Moreover weintroduced the natural join and SVM based feature selection in multi-relation statisticallearning. The performance of our model on various datasets indicates that our model isefficient, reliable and highly accurate.
Support Vector Machine; Natural Join; Statistical Learning; Multi-relation; Inductive Logical Programming; Clusters; Feature Selection
Full Text: PDF (downloaded 465 times)
- There are currently no refbacks.