4- Course Content :-

4.1- Fundamentals of pattern recognition. Concepts, terminology, evaluation of algorithms.

4.2- Bayes theory, normal density, discriminant analysis for normal density, error probability.

4.3- Density estimation: Parzen windows, k nearest neighbor, Probabilistic neural networks.

4.4- Feature extraction and selection: Principal component analysis (PCA), Fisher’s linear discriminant and other linear approaches.

4.5- The perceptron model – multiplayer perceptron neural networks, gradient descent optimization.

4.6- Kernel Methods Introduction.

4.7- Kernel methods – Part II: Support Vector Machines.

4.8- Other Kernel Methods – Relevance Vector Machines, Kernel PCA.

4.9- Other kernel based approaches, mixture of Gaussians, EM algorithm.

4.10- Multiple Classifier Systems. Concept of diversity, ensemble creation algorithms I – Bagging, boosting and AdaBoost, Bias – variance dilemma.

4.11- Applications of ensemble systems II – Feature selection, missing feature, error correcting output codes, learning in nonstationary environments.

4.12- Unsupervised Learning, clustering algorithms, self organizing maps.

4.13- Decision Trees, CART algorithm.

4.14- Hidden Markov ModelsSource localization and spectral estimation.

4.15- Application traffic classification.

4.16- Distributed Hash Tables (DHT).

4.17- Overlay networking.

4.18- Peer-to-peer networks and security.

4.19- Privacy and anonymity.

4.20- Internet Games.

4.21- Network geography.

4.22- Network measurement.

4.23- Distributed data streaming.

4.24- TCP.

4.25- Future directions for the Internet.