Department of Information & Communication Systems Engineering
University of the Aegean
SCHOOL OF ENGINEERING

Department of Information
& Communication Systems Engineering

Information & Communication Systems Security
Information Systems
Artificial Intelligence
Computer & Communication Systems
Geometry, Dynamical Systems & Cosmology
 


Title:
Lesson Code:
Semester:
ECTS:
Theory Hours:
Lab Hours:
Faculty:
 
Content outline
Pattern Recognition Systems - Prediction Problems - Patterns and Features - Classifiers - Bayes classifier, Likelihood Ratio, Error Probability , Cost, Risk Bayes - Non-Parametric Density Estimation - Histogram, Parzen Windows , smoothed Kernel, density estimation by kNN - The classification rule of k nearest neighbor (k-NN) - the curse of dimensionality - Feature extraction - Features Selection , Principal Component Analysis PCA, Linear Discriminant Analysis LDA - Filters, Serial algorithms, exponential algorithms, Random algorithms - deviation and variance, holdout, Cross Validation, Bootstrap - unsupervised learning - Mixture models, EM algorithm - Non-parametric unsupervised learning - Proximity Measures, algorithms k-means, ISODATA, Hierarchical clustering, dendrogram - SVM - HMM.
 
Learning outcomes
The course intends to familiarize the students with Pattern Recognition Systems, known methodologies and applications of the field. Moreover, it demonstrates how those systems are designed, developed and evaluated. On a second level, the student is asked to design and develop their own Pattern Recognition System for specific patterns or application.
 
Prerequisites
Not required.
 
Assessment/Grading Methods
 
Language of Instruction
Greek, English (for Erasmus students)
 
Μode of delivery
Face-to-face


Home | Contact

University Of The Aegean

SCHOOL OF ENGINEERING
Department of Information & & Communications Systems Engineering

© Copyright ICSD :: 2008 - 2017