Building a highly accurate prediction model is certainly a difficult task. Noise – Irreducible error i.e. the part of target value which the model is not able to predict / explain. As you know it is impossible to reduce the noise, hence the term irreducible error, we shift our focus on reducing Bias and Variance. So, Ensemble learning methods bring up the technique to reduce the Bias and Variance of the model by using multiple models together (hence the term Ensemble), in order to achieve better predictive performance, instead of a single model for prediction. There are a number of Ensemble methods, in this article I will be discussing about the two widely used Ensemble methods that are Bagging and Boosting. When we use different / single learning algorithm, multiple times for prediction.

Bagging Bootstrapping The Bagging Algorithm Why Bagging Works LAB: Bagging Models Data scientist is called as the sexiest job of the 21st century. They take an enormous mass of messy data points (unstructured and structured) and use their formidable skills in math, statistics, and programming to clean, massage and organize. But worry not we are here to the rescue and teach you how to be a data scientist, more importantly, upgrade your analytic skills to tackle any problem in the field of data science. Join us on "statinfer.com" for becoming a "scientist in data science" Our "Machine Learning" course is now available on Udemy https://www.udemy.com/machine-learnin... Facebook link:- (Visit our facebook page we are sharing data science videos) https://www.facebook.com/aboutanalytics/ Visit our official website to go deeper into data science topics.

Abstract--In Machine Learning, ensemble methods have been receiving a great deal of attention. Techniques such as Bagging and Boosting have been successfully applied to a variety of problems. Nevertheless, such techniques are still susceptible to the effects of noise and outliers in the training data. We propose a new method for the generation of pools of classifiers based on Bagging, in which the probability of an instance being selected during the resampling process is inversely proportional to its instance hardness, which can be understood as the likelihood of an instance being misclassified, regardless of the choice of classifier. The goal of the proposed method is to remove noisy data without sacrificing the hard instances which are likely to be found on class boundaries. We evaluate the performance of the method in nineteen public data sets, and compare it to the performance of the Bagging and Random Subspace algorithms. Our experiments show that in high noise scenarios the accuracy of our method is significantly better than that of Bagging. Ensemble methods [1] [2] [3] are techniques that combine multiple predictors trained independently, using a combination of the outputs of each predictor as the final output. This is in contrast to traditional Machine Learning methods, which train a single classifier on the whole of the training set.