2 research outputs found

    CARBO: Clustering and rotation based oversampling for class imbalance learning

    No full text
    Class imbalance of a data set is a crucial problem in machine learning where one class significantly outnumbers others. In such a data set, classification is a troublesome task for the standard classification algorithms, leading to bias towards the majority class. Different methods have been developed so far, such as oversampling, undersampling, and cost-sensitive learning, to deal with class imbalance circumstances. Among these techniques, oversampling technique does not suffer from the information loss and critical cost selection challenges. However, appropriate synthetic sample generation can be challenging and vulnerable to privacy leakage. This research proposed an oversampling technique, called CARBO, using threshold-based geometric rotation and majority class influenced clustering. Unlike the existing resampling approaches to class imbalance problem, we contribute to consider the data privacy and optimal sample generation together for effective oversampling. The performance of CARBO is evaluated using 44 benchmark imbalanced data set. The empirical analysis elucidates that CARBO can make boosting-based C4.5 ensemble classifiers perform higher for 73% of the data set than six state-of-the-art approaches. In addition, the theoretical compatibility analysis of CARBO demonstrates its robustness.No Full Tex

    Studies on the genetic diversity of pointed gourd using biochemical methods (<i>Isozyme analysis</i>)

    No full text
    Biochemical characterizations of 64 pointedgourds were done using three isozyme viz., acid phosphatase, peroxidase and glutamate oxaloacetate transaminase. A wide range of diversity among the gremplasm based on their acid phosphatase, peroxidase and glutamate oxaloacetate transaminase isoenzyme banding patterns were observed. In respect of isoenzyme activity; 8 acid phosphatase, 7 peroxidase and 9 glutamate oxaloacetate transminase electrophoretic zymotypes were formed by 19, 11, and 19 bands at different Rf values varying from 0.19 to 0.82, 0.38 to 0.69 and 0.15 to 0.95, respectively. The wide range of similarity co-efficient of 0.0-80.0, 0.0-66.0, and 0.0-80.0 as found among the electrophoretic patterns in acid phosphatase, peroxidase, and glutamate oxaloacetate transminase, respectively, indicating wide genetic diversity among the accessions. Based on the polymorphic activity of these three enzymes, 27 combinations of electrophoretic zymotypes were identified, each of which can he equated to genotypes. Each of the groups consisted of one to eight genotypes. Sixty four accessions of pointed gourd were grouped into 12 clusters. The genotypes collected from the same location were grouped into different clusters. Key Words: Genetic diversity; pointed gourd; biochemical methods. DOI: 10.3329/bjar.v34i1.5762Bangladesh J. Agril. Res. 34(1) : 123-141, March 2009</jats:p
    corecore