Support vector machines (svm)
- 2. Support Vector Machines (SVM) - Supervised learning methods - Associated learning algorithms - They can represent linear & non-linear functions and they have an efficient training algorithm SS_PATIL12-02-2020 2
- 4. There are 2 kinds of SVM classifiers: 1.Linear SVM Classifier 2.Non-Linear SVM Classifier SS_PATIL12-02-2020 4
- 5. • Example: Fig. 1 Fig. 2 SS_PATIL12-02-2020 5
- 7. Example of Bad Decision Boundaries SS_PATIL12-02-2020 7
- 8. Tuning Parameters Kernel Regularization Gamma Margin. SS_PATIL12-02-2020 8
- 9. Kernel linear kernel f(x) = B(0) + sum(ai * (x,xi)) Polynomial kernel K(x,xi) = 1 + sum(x * xi)^d exponential Kernel K(x,xi) = exp(-gamma * sum((x — xi²)). SS_PATIL12-02-2020 9
- 10. Regularization To avoid misclassifying each training example low regularization value high regularization value SS_PATIL12-02-2020 10
- 12. Margin separation of line to the closest class points. SS_PATIL12-02-2020 12
- 13. Problem Statement: Use Machine Learning to predict cases of breast cancer using patient treatment history and health data. SS_PATIL12-02-2020 13
- 14. Dataset: Breast Cancer Wisconsin (Diagnostic) Dataset Let us have a quick look at the dataset SS_PATIL12-02-2020 14
- 15. Step 1: Load Pandas library and the dataset using Pandas Step 2: Define the features and the target SS_PATIL12-02-2020 15
- 16. Have a look at the features: SS_PATIL12-02-2020 16
- 18. Step 3: Split the dataset into train and test using sklearn before building the SVM algorithm model Step 4: Import the support vector classifier function or SVC function from Sklearn SVM module. Build the Support Vector Machine model with the help of the SVC function SS_PATIL12-02-2020 18
- 19. Step 5: Predict values using the SVM algorithm model Step 6: Evaluate the Support Vector Machine model SS_PATIL12-02-2020 19
- 20. SVM Applications: • SVMS are a by product of Neural Network. They are widely applied to pattern classification and regression problems. Here are some of its applications: • Facial expression classification • Speech recognition • Handwritten digit recognition SS_PATIL12-02-2020 20
- 21. Advantages of SVM Classifier: More effective It works effectively Memory efficient It is a robust model SS_PATIL12-02-2020 21
- 22. Disadvantages of SVM Classifier: • Not suitable • It does not perform very well • good generalization performance • SVMs have high algorithmic complexity and extensive memory requirements due to the use of quadratic programming. SS_PATIL12-02-2020 22