That sounds good, will you be querying the blockchain directly or using HiveSQL?
I had a look at children (replies/comments) on posts since hive, the order of the list is the average number of children per post.
Let me know if I can be of any help, cheers.
I have my own python scripts using the Beem library developed by @holger80. I still need to finalise that but I was busy with writing a research paper on detector simulations (it is now finished so that I could have again time for other things).