Relationship Mining in PISA CBA 2012 Problem Solving Dataset Using Association Rules

The aim of this work was to examine whether the background variables can be linked to the problem solving assessment performance in PISA 2012 data by applying data mining techniques. Data from two countries, Hungary and Finland were processed by association rules mining. Interesting rules were extracted by post-processing of the rules got by all the mining processes for five plausible variables. Resulting rules are outlined with their corresponding confidence values. Confidence values, and support values were computed and depicted in a scatter plot.