Semi-supervised Nonlinear Distance Metric Learning via Random Forest and Relative Similarity Algorithm

1 Research Scholar, Department of Computer Science, Vellalar College for Women, Erode, Tamilnadu, India 2 Assistant Professor, Dept. of Computer Applications, Vellalar College for Women, Erode, Tamilnadu, India ---------------------------------------------------------------------***--------------------------------------------------------------------Abstract Similarity measure is closely related to distance metric learning. Metric learning is the task of learning a distance function over objects. In the base work, a nonlinear machine learning method is implemented by using Semi-Supervised Max-Margin Clustering to construct a forest of cluster hierarchies. In that individual component of the forest represent cluster hierarchies. Clustering hierarchies gives handling any form of similarity or distance. It is also used for applicability to any attributes type. Most hierarchal algorithms do not revisit once constructed clusters with the purpose of improvement. For distance metric learning give some computational complexity. To reduce the complexity and improvement purpose, proposed algorithm called Relative Similarity use the linear reconstruction weights to measure the similarity between the adjacent points. The original data points are collected in dimensional space and the goal of the algorithm is to reduce the dimensionality. The proposed algorithm gives good clustering results and reasonably fast for sparse data sets of several thousand elements. The investigational tests prove that the leads to enhanced performance than the presented approach by analyzing the code quality in an efficient manner.