Information Fusion(Ensemble System)in Data Warehousing and Data Mining: Ensemble System(Multiple Classifier System) In Data Mining - Tapa blanda

Information fusion is becoming a major need in Data Mining. Typical applications of these techniques include data modeling (ensemble methods). The behavior of various classification algorithms differs based on accuracy and computational complexity. For some algorithms there may be a significant variation in the performance when some parameters are varied. In this research the behavior of the modified AdaBoost algorithm with NN as a base classifier and as a preprocessing step feature selection combined with the evaluation schemas (like subset evaluation, consistency based, correlation based, filter approach, wrapper approach etc.) are applied by varying the number of parameters. Predictive accuracy is substantially improved when combining multiple predictors. A novel idea of an Ensemble System applying Boosting to Neural Networks for High Dimensional Datasets. The method uses Genetic Algorithms (to select relevant features) for essential feature selection with various Evaluation Schemes. As Genetic Algorithms deal well with large solution spaces, tuning it to adjust as per the requirements of the ensemble, we can get optimum feature selection. Finally Boosting algorithm that finishe

Amit P. Ganatra (B.E.-’00-M.E. ’04-Ph.D. ’12) He is concurrently holding Professor (Jan 2010 till date), Headship in computer Engineering Department (since 2001 to till date) at CSPIT, CHARUSAT and Deanship in Faculty of Technology-CHARUSAT, Gujarat (since Jan 2011 to till date).