Outlier analysis and Detection using K-medoids with support vector machine

Abstract

Spatio -temporal methods is the process of innovations and finding the patterns from the knowledge representations through outliers. This kind of data representing the (i) the states of an object (ii) position or event in space at a particular period of time. It refers to the Objects whose attribute values are entirely different from its neighbourhood. Always their locations are different even the nodes from the entire population are unique. Outlier Detection is the most important techniques in data mining, which is useful for identifying several activities from the huge data set. This Project is deals with the identification of Breast cancer. Here we are comparing the accuracy and performance with the previous technology, as expected Our proposed algorithm using k-medoids -support vector machine is more accurate then the Rough Outlier set Extraction mode.

Custom Outlier Detection for Electrical Energy Consumption Data Applied in Case of Demand Response in Block of Buildings

“…Neural networks [ 42 , 43 , 44 ] and support vector machines [ 45 ] were also used for anomalous data detection. Functional dependency thresholding with Bayesian optimization for functional dependencies data cleaning purpose was successfully tested on synthetic and real data [ 46 ], relaxed functional dependencies were also detected using an improved discovery algorithm relying on a lattice structured search space with new pruning strategy [ 47 ].…”

A Fuzzy Density Peak Optimization Initial Centers Selection for K-medoids Clustering Algorithm

“…K-medoids algorithm is a good solution to this problem, the algorithm does not use the average value of the objects in the cluster (center of mass) as the reference point, but the representative object is called the center point instead of center of mass . K-medoids algorithm has been applied to many fields such as the outlier analysis, the detection [3] and the distributed computing [4] etc.…”