Online Anomaly Detection under Over-sampling PCA
Journal Title: International Journal for Research in Applied Science and Engineering Technology (IJRASET) - Year 2014, Vol 2, Issue 12
Abstract
Anomaly detection is the process of identifying unusual behavior. Outlier detection is an important issue in data mining and has been studied in different research areas. In this paper we use “Leave One Out” procedure to check each individual point the “with or without” effect on the variation of principal directions. Based on this idea, an over-sampling principal component analysis (osPCA) outlier detection method is proposed for emphasizing the influence of an abnormal instance. Except for identifying the suspicious outliers, we also design an online anomaly detection to detect the new arriving anomaly. In addition, we also study the quick updating of the principal directions for the effective computation and satisfying the online detecting demand. It is widely used in data mining; the proposed framework is favored for online applications which have computation or memory limitations. Compared with the all existing algorithms, our proposed method is in terms of flexibility, accuracy and efficiency.
Authors and Affiliations
P. Chinababu, Mrs. A. Vanathi
Keywords
Related Articles
- EP ID EP19178
- DOI -
- Views 274
- Downloads 4
Modular Cascaded H-Bridge Multilevel PV Inverter by Using Fuzzy Logic Controller with Distributed MPPT for Grid-Connected Applications
This paper presents a modular cascaded H-bridge multilevel photovoltaic (PV) inverter for single- or three-phase gridconnected applications. The modular cascaded multilevel topology helps to improve the efficiency and fl...
Experimental Investigation on Effective Utilization of Natural Wastes with Lime for Making Light Weight Bricks
In this modern era considerable attention has been paid to the utilization of alternative materials, which bear higher engineering quality than traditional materials and are financially affordable. With that in mind, a...
Study on the Properties of Concrete by Using Super Absorbent Polymer
The use of super absorbent polymer (SAP) in concrete is confirmed to have many constructive effects on the properties of concrete in its both stages; fresh concrete and hardened concrete. This study focus on the use of...
Risk Assessment of Heavy Metal Toxicants in Hair Oils
The hair oils were analyzed for heavy metal poisoning using Double Beam UV-Visible spectrophotometer and AAS. The hair oils were procured from the market and analyzed for heavy metals viz. Pb, Cu, Fe, Ni, Cr, Zn and Mn....
Design and Thermal analysis of Piston in Multi Cylinder Engine by Interring Gas Dynamics
The piston ring performs its function as a seal of high and low pressure sides in a cylinder and a medium of heat transfer from piston to cylinder walls. The operation of piston ring packs influences the performance, ef...