Items list

Subject Code: IC469
ECTS: 6
Study Program: Undergraduate Degree Program
Teaching Semester: Semester 7
Type
Επιλογής
Κατεύθυνση ΓΝΜ: Ελεύθερης Επιλογής
Κατεύθυνση ΘΡΜ: Ελεύθερης Επιλογής
Κατεύθυνση ΕΦΜ: Ελεύθερης Επιλογής
Κατεύθυνση ΠΛΗ: Ελεύθερης Επιλογής
Κατεύθυνση ΣΠΕ: Ελεύθερης Επιλογής
Teaching Hours (Theory / Tutorial / Laboratory): 2-0-2
Erasmus Students: No
Description
Data description with graphs and tables. Presentation of the basic statistical measures for data description. Data Preparation. The importance of data control and "cleaning" (data cleaning). Introduction to Databases. SQL Introduction to supervised learning: decision trees, logistic regression. Introduction to regression: Multiple linear regression. Forecasting. Improving a model. The problems of over-parametrization. Model performance control. Dimensionality Reduction. The feature selection process. The Principal Component Analysis method with SVD factorization of matrices. Unsupervised learning, Clustering. Applications and evaluation of k-means. Application of Hierarchical Clustering models. Semi-supervised learning. Introduction to metadata and Big Data. Computational Methods for Big Data Analysis (Hadoop and MapReduce). Lab: (i) Introduction to the R language for Data Science. (ii) Creating, selecting and comparing categorical data using Factors. Storing data tables in Data Frames. Selecting elements from a Data Frame and converting to a Table. (iii) Basic graphics/visualization packages in R. (iv) Functions - Loops - Flow control. (v) Introduction to SQL. Queries. Queries on multiple tables with the JOIN command. Operators. Subqueries. (vi) Rattle. (vii) R Hadoop. Information: Replaces the "Databases" Course from the Academic Year 2018-19.