Optimization of Regression algorithms using Learning curve in WSN

Regression algorithm used for the prediction of output with given features and it is a supervised learning algorithm. In applying regression algorithms such as linear regression, Regression using ANN, Regression using deep learning there comes a problem of (1) High bias or under fitting due to simple model or (2) High Variance due to complex model. High variance problems occur when we take high order polynomial in linear regression, more hidden layers in ANN, or Deep Learning. So to find which types of problems occur i.e. High bias or variance, one has to use the learning curve to find an acceptable error. This paper will be helpful for those researchers who are working in the regression algorithm and have lesser knowledge in the mathematical treatment of the regression algorithm. In this, simulation has done with the GNU octave simulator and verified the concept of the learning curve.