if in a model i used a weak learner decision tree with an ensemble method adaboost for increasing accuracy.
Q1: what will be called this model. Is it dependent ensemble model or other?
Q2:How dependent ensemble model works during making a weak learner to strong learner?
saqlain
Ensembling in machine learning is the process of training multiple learners using either same or different approach in order to leverage their weaknesses; the results of the learners are then combined in order to improve their accuracy and efficiency. Popular ensembling approaches are Boosting, Bagging and Stacking.
1. Since you are boosting a weak model with another (Adaboost), then any model that results is dependent on the model used to boost
2.These links will be helpful for the second question on ensembling techniques (b) and overview of implementations(a)
(a). http://machinelearningmastery.com/machine-learning-ensembles-with-r/
(b). https://www.analyticsvidhya.com/blog/2015/08/introduction-ensemble-learning/
You need a family of base models ( weak or strong) to build an ensemble model. The family can contain models of same type as in your case of adaboost, or of different type.(linear , neural ,svm etc).
Ensemble-based approach is basically used to boost performance of traditional methods. The ensemble-based approach is based on the recently emerged technique known as boosting. Generally, boosting-like learning rules are not suited to a strong and stable learner. The idea behind boosting is to sequentially employ a base classifier on a weighted version of the training sample set to generalize a set of classifiers of its kind. Often the base classifier is also called learner. Boosting is particularly robust in preventing over-fitting and reducing the generalization error by increasing the so-called margins of the training examples.
You can also check the link below for more clarifications of your question
https://www.toptal.com/machine-learning/ensemble-methods-machine-learning
- Badges
- Science topic