| Block I: INTRODUCTION TO INTELLIGENT SYSTEMS. |
Topic 1: FUNDAMENTALS
Motivation, Basic principles, Approaches, Learning, Applications.
Topic 2: CLASSIFIER DESIGN CYCLE
Data collection, Feature selection, Model selection, Classifier tuning/training, Evaluation (metrics and techniques).Naive Bayes classifier.
Topic 3: PARAMETRIC SUPERVISED LEARNING TECHNIQUES.
Linear and logistic regression.
Topic 4: NON PARAMETRIC SUPERVISED CLASSIFICATION TECHNIQUES.
Fundamentals. Neighborhood classifiers.
Topic 5: MODEL EVALUATION AND SELECTION.
Estimation techniques: cross-validation, hold-out, bootstrap, model comparison, classification performance metrics, other predictive model metrics: quadratic error, standard error.
Topic 6: ARTIFICIAL NEURAL NETWORKS.
Introduction, Topology and activation functions, The simple perceptron, The Adaline network, Multilayer perceptron (MLP).
Topic 7: DATA PREPROCESSING.
Anomalous data detection. Missing samples. Data integration and normalization. Data transformation. Feature selection and extraction.
Topic 8: UNSUPERVISED LEARNING.
Introduction, Space, distance and similarity, Partitioning and agglomerative techniques, Method of centroids (K-means).
Topic 9: OTHER CLASSIFICATION MODELS.
Support Vector Machines (SVM). Decision trees.
Report - Practical applications
Application of Intelligent Systems in Bioinformatics, Information retrieval systems, Spam detection, Fake news detection, Inappropriate content detection, Music analysis, Sentiment analysis and opinion data, Biometric recognition, Manufacturing environments, Industrial and environmental environments, Chemometrics, Intelligent laboratories, Cancer prediction models, Cybersecurity. |
