Pattern Recognition (for AI)
Projects
Project 1
- Task 1: Pairwise correlation coefficients
- Task 2: Hamming distance
- Task 3: Covariance matrix
- Task 4: 2D Gaussian
- Task 5: Naive Bayesian rule
- Task 6: Decision Tree
- Task 7: Learning Vector Quantization
- Task 8: Cross-Validation
- Task 9: ROC
- Task 10: Transforms
- Task 11: K-Nearest Neighbor
- Task 12: K-means clustering
Project 2: Creating Pattern Recognition Pipelines using Traditional Machine Learning methods
Description
This course provides an introduction to the theory and practice of pattern recognition. It is the research area that studies the design and operation of systems that detect, identify, recognise or classify patterns in data. Important application domains are image analysis (e.g. licence plate recognition or various medical applications), computer vision, speech analysis, man and machine diagnostics, person identification (e.g. by iris or fingerprint), spam filtering, industrial inspection, financial data analysis and forecast, genetics. Generally, pattern recognition includes techniques such as feature extraction, classification, and error estimation. The course presents various classification techniques, e.g. k-nn, LVQ, SVM, decision tree, ANN, CNN, GAN, and clustering, e.g. k-means, VQ, dendrogram, gap statistics. Various applications are presented throughout the course.