Level of study: 3.
Year: 1.
Compulsory subject
The aim of the subject is to provide students with knowledge in the field of data economy. The course deals with the analysis and processing of large data sets (Big Data). Students should become familiar with the creation, structure, and management of data warehouses, understand approaches to acquiring knowledge from data, and gain advanced knowledge of neural networks.
Brief course outline:
- Big data - Big data, the complexity of big data, big data processing architectures, big data technologies, the commercial value of big data, data warehouses, data warehouse management, workflow management in data warehouses,
- Data mining - Data and file formats (structured, unstructured, etc.), SQL and databases, text processing (parsing, tokenizing, stemming, etc.),
- Data representation (representation of feature vectors, etc.) The need for data mining, Data preprocessing: dimension reduction, missing value analysis, normalization and standardization, noise and outlier detection.
- Pattern detection, classification, association, and prediction techniques. Machine learning Basic concepts of neural networks, characteristics of neural networks, terminology, application of neural networks. Supervised machine learning, unsupervised machine learning, reinforcement learning.
- Knowledge discovery, artificial intelligence, learning rules, error correction learning, memory-based learning, Hebbian learning, competitive learning, Boltzmann learning, single-layer perceptron, multi-layer perceptron, backpropagation, recurrent networks, network simplification, adaptive networks,
- Decision-based neural networks, hierarchical neural networks, probabilistic neural networks, radial basis function networks, comparison of RBF and multilayer perceptron networks.
- Classification of linearly separable patterns, Boltzmann machine, Helmholtz machine, Support vector machines, Self-organizing migration algorithm, genetic algorithms, prediction systems.
