Data Mining

A hidden challenge in this work is to to work with real-life data and real-time performance analysis and comparison. The running time for certain dataset for a particular threshold should be too high that may not end in years. On the other hand for threshold on a particular dataset, there may not be any patterns available. So, choosing right threshold for each dataset is a challenge. In order to save time from waiting for outputs, I wrote a batch file for windows or a shell script for linux/mac where all the runs with required parameters or switches are listed. Once I ran the script or batch file from command line, I might set out for other works leaving my computer dealing with the computation by itself. Just we have to make sure no one turns computer off while the batch/script is running. All kinds of exception handling are included in the program. In case of an invalid switch or invalid value for a particular switch, program should show a helpful hint instead of treating the value in an inappropriate way or terminating abruptly.