In a current problem we are trying to investigate the impact of global properties of developer social networks (made based on comments in the bug report; vertices are developers and edges exist if two developers comment on same bug report) on bug fixing process.

We want to see the evolution of such network over the period of six months and how the global properties like clustering coefficient or modularity etc. are related to bug fixing process attributes like the number of bugs fixed in those six months. In this way we have many sets of pairs of independent and dependent variables. But the number of pairs in each set is just 16, i.e. we have 16 rows having two columns one column represents the independent variable like clustering coefficient of network formed in a 6 month snap shot and the other column has dependent variables, like the number of bugs fixed in the same snap shot.

More Avdhesh Gupta's questions See All
Similar questions and discussions