Ce diaporama a bien été signalé.
Le téléchargement de votre SlideShare est en cours. ×

Ensemble Method.pptx

Publicité
Publicité
Publicité
Publicité
Publicité
Publicité
Publicité
Publicité
Publicité
Publicité
Publicité
Publicité
Prochain SlideShare
BaggingBoosting.pdf
BaggingBoosting.pdf
Chargement dans…3
×

Consultez-les par la suite

1 sur 18 Publicité

Plus De Contenu Connexe

Similaire à Ensemble Method.pptx (20)

Plus récents (20)

Publicité

Ensemble Method.pptx

  1. 1. Analytical Method It is a generic process combining the power of the Scientific Method with the use of formal process to solve any type of problem. It has these nine steps: 1. Identify the problem to solve. 2. Choose an appropriate process. 3. Use the process to hypothesize analysis or solution elements. 4. Design an experiments to test the hypothesis. 5. Perform the experiments. 6. Accept, reject, or modify the hypothesis. 7. Repeat steps 3, 4, 5, and 6 until the hypothesis is accepted. 8. Implement the solution. 9. Continuously improve the process as opportunities arise.
  2. 2. Analytical approach Analytical approach is an appropriate process to break a problem down into the elements necessary to solve it. Each element becomes a smaller and easier problem to solve. According to Morgan Jones, former CIA analyst, An analytical approach is also known as "structuring one's analysis.”
  3. 3. ENSEMBLE METHODS Ensemble methods are used to create stronger (More Accurate) classification tree models. This is done by combining weak classification tree models to create stronger versions.
  4. 4. ENSEMBLE METHODS XLMiner offers three robust ensemble methods: • Bagging • Boosting • Random Trees
  5. 5. ENSEMBLE METHODS These methods differ in terms of :  How data is selected from the weak dataset?  How the weak models are generated?  How the outputs are combined to form a stronger classification tree model? Specific ensemble methods may be more appropriate in certain situations based on the nature of the dataset (e.g., size, ability to parallelize).
  6. 6. ENSEMBLE METHODS However, by using ensemble techniques, we can reduce the bias and variance of a model. This reduces the overall generalization error. • Bias: Bias is an error which arises due to false assumptions made in the learning phase of a model. • Variance: Variance tells us how sensitive a model is to small changes in the training data.
  7. 7. Bagging Bagging (bootstrap aggregating) is a simple but powerful ensemble algorithm that facilitates the increased stability & accuracy of classification models. It works by generating multiple training datasets via random sampling with replacement (RSR), applying the algorithm to each dataset, and then taking the majority vote amongst the models to determine data classifications.
  8. 8. Bagging
  9. 9. Bagging Advantages Reduces variance. Helps to prevent over fitting. Easily parallelized for application to large datasets.
  10. 10. Boosting Boosting facilitates the conversion of weak learners (i.e. classifiers with weak correlations) to strong learners (i.e. High correlated classifiers). Boosting creates strong classification tree models by training models to concentrate on misclassified records from previous models; when this is done, all classifiers are combined by a weighted majority vote. This process places a higher weight to incorrectly classified records while decreasing the weight of correct classifications. This effectively forces subsequent models to place a greater emphasis on misclassified records. The algorithm then computes the weighted sum of votes for each class and assigns the best classification to the record.
  11. 11. Boosting
  12. 12. Boosting Disadvantages It is not capable for parallelization. Advantages reduces bias and variance.
  13. 13. Differences Between Bagging and Boosting
  14. 14. Random Trees Random Trees borrows the Bagging concept of random feature selection to construct decision tree models with controlled variance. The Random Tree method uses training datasets to create multiple decision trees and then using the mode from each class to create a strong classifier.
  15. 15. Random Trees Advantages reduces variance. helps to prevent over fitting. easily parallelized for application to large datasets. The generation of models is faster than Bagging.
  16. 16. How to Access Ensemble Methods in Excel? • Launch Excel. • In the toolbar, click XLMINER PLATFORM. • In the ribbon's Data Mining section, click Classify. • In the drop-down menu, hover the mouse cursors over Classification Tree to reveal a sub-menu. • Select Boosting, Bagging, Random Trees, or Single Tree as needed.

×