Outlier detection in high dimensional data based on elastic net regression

Citations of this article
Mendeley users who have this article in their library.
Get full text


Outlier detection in large datasets is the dynamic research area in computer science such as data mining, database systems, and distributed systems. Outlier detection faces many challenges due to the absence of data samples from the outlier class. Massive algorithms have been projected to conquer the challenges in this field to improve the efficiency of regression approach for large datasets. Currently, no particular efficient regression technique is designed for outlier detection. In this research, we proposed an ElasticNet regression model for detecting the outliers in high dimensional data. To validate the efficiency and competence of our projected algorithm, it is implemented in the open source software called Weka Explorer. The parameters such as Mean absolute error 0.0022, RMSE 0.0387, Relative absolute error (RAE) 0.4562 and Root relative squared error (RSE) 7.8722 are calculated using annthyroid dataset. ElasticNet model consumes less computational time, generates fast convergence results, provides high accuracy and correctly classified accuracy is 98.25%.




Anuradha, C., Ramesh, M., & Murty, P. S. R. C. (2019). Outlier detection in high dimensional data based on elastic net regression. International Journal of Innovative Technology and Exploring Engineering, 8(12), 325–328. https://doi.org/10.35940/ijitee.L3479.1081219

Register to see more suggestions

Mendeley helps you to discover research relevant for your work.

Already have an account?

Save time finding and organizing research with Mendeley

Sign up for free