Regression Models and Knowledge Processing

Description

Knowledge Data Processing with statistical methods combined with learning processes, involves:

  1. Data organization for small and big data sets [Bratsas C.]
  2. Descriptive statistics using R [Bratsas C.]
  3. Analysis of Interdependence with Correlation, Regression, Singular Value Decomposition, Principal Components Analysis, Neural Networks, Link Identification. [Antoniou I.]
  4. Discrimination with Proximities (Affinities-Similarities). Mutual Information. [Antoniou I.]
  5. Analysis of Variance (ANOVA)  using R[ [Bratsas C.]
  6. Categorization-clustering and Classification,  (Partitional and Hierarchical) with Decision Trees, Artificial Neural Networks, Nearest Neighbor Classifiers, Logistic Regression, Bayes Networks, Support Vector Machines. (examples in R) [Bratsas C.]
  7. Interpretation of Results, evaluation of classification models (cross validation, Split test methods, accuracy and precision, recall, area under ROC curve). Representation and visualization of data and of results (examples in R) [Bratsas C.]
  8. Αpplications to real data and Big Data [Bratsas C.]

Laboratory Software: Open source R (R-studio)

Suggested References

  1. http://www-bcf.usc.edu/~gareth/ISL/ISLR%20First%20Printing.pdf
  2. http://statweb.stanford.edu/~tibs/ElemStatLearn/
Semester: 
Units: 
4
Credit Units (ECTS): 
5.5
Hours: 
4ώρες
ID: 
0531
Course Type: 
X