Online social networks are popular habitats for many Web users. Research on activity patterns of individual features of online social networking systems is ongoing. Our approach on the study of these patterns is more encompassing than previous efforts. We have created a modern experimental educational online social network for the purpose of the study of network structures and communication phenomena in these websites. In this paper presented are patterns of activity usage of most classic features of online social networks: friends, private massages, forum, blog, photogallery, comment wall. Activity patterns of these features are compared against each other, also by taking into account the main user attribute.

Complexity of data analysis in data mining often makes results difficult to interpret. This problem could be solved using various approaches. Principal Component Analysis (PCA) and Disjoint Cluster Analysis (DCA) are methods used for data reduction and summarization. In this paper, PCA and DCA were applied on dataset example containing information about students' courses and time necessary to pass related exams. The SAS software was used as a data mining tool for performing this analysis. Another approach for better interpretation is visualization of results. This means showing important attributes visually to aid informal users to interpret results.

No abstract

No abstract

No abstract