• open the file
• skip the first two line
• while there is more data in the file:
  • split each line in two parts, a unique region identifier a data type and possibly a record count.
  • store the last two values in a hash of hashes keyed on region and column name
• Iterate over the keys and do whatever it is you want to do with them

Have a go at coding that up, if it doesn't work for you, bring your issue to the forum and we'll do our best, but we are a teaching monastery, not a code writing service