Critical Nuggets Identification for Classification Task | Abstract

ISSN ONLINE(2320-9801) PRINT (2320-9798)

Special Issue Article Open Access

Critical Nuggets Identification for Classification Task


Pattern detection and outliers has emerged as an important area of work in the field of data mining. Critical nuggets are small collections of records or instances that contain domain specific important information for classification. The system identifies the critical nuggets to measure the criticality of data instances. Critical nuggets are identified using CRscore which helps in improving classification accuracies in real-world data sets. The CRscore values are stored in histogram and highest score values are identified. Using the highest CRscore the critical nugget is identified in the data sets. It improves the accuracy of classification. Class imbalance problems have drawn growing interest recently because of their classification difficulty caused by the imbalanced class distributions.


To read the full article Download Full Article