18C5T13 Study Guide - Final Guide: Unit, Hierarchical Clustering, Binary Tree

• School: Jawaharlal Nehru Technological University - JNTU Kakinada
• Department: Engineering
• Course: 18C5T13

Document Summary

Unit -iv:linear models: the least-squares method, the perceptron: a heuristic learning algorithm for linear classifiers, support vector machines, obtaining probabilities from linear classifiers, going beyond linearity with kernel methods. distance based models: introduction, neighbours and exemplars, Nearest neighbours classification, distance based clustering, hierarchical clustering. Linear models are stable(limited impact on learned model) Linear models are less likely to overfit the training data. Labelled data is called linearly separable if there exists a linear decision boundary separating the classes. The least-squares classifier may find a perfectly separating decision boundary if one exists, but this is not guaranteed: now, move all but one of the positive points away from the negative class. The decision boundary will also move away from the negative class, at some point crossing the one positive that remains fixed. The perceptron iterates over the training set, updating the weight vector every time it encounters an incorrectly classified example.