Here is an example of how to apply this method. From a birth and death registry of say twenty million people, mine the data to see if any planetary alignments at birth or death consistently match longer or shorter life spans. Approach the data without a specific hypothesis except that correlations may exist, with the aim simply to look to see what is there. Such correlations have eluded searches at the 'kilobyte' level but might be amenable to data mining at the petabyte level.