Data Analysis

I’ll describe here how I image the data analysis process will look like:

Get the Data:

• Every hour the system will try to get a more recent data from WOW API on the auctions of all registered Realms;
• This auction’s data will be saved with the timestamp of the last_modified field from WOW API;

Analysis:

When a user, ask for the tracked item (ex: #12345), it will:

• get the realm it’s been asked for;

• get the datetime for the latest 30 days, counting on today;

• For each day, it will first try to check if the information is already saved in a Django Model (PostgreSQL), this will happen if the analysis for that day/realm/item was already done;

• if the information is not present in PostgreSQL, then search the MongoDB’s data and get all the auctions.json of that day;

  • after that, it will go through each auctions.json, and get the Avarage Price and the User Price for the tracked item (ex: #12345);
  • then it will make the avarage for the hole day with the Avarage Price of each auction.json, and the same thing with the User Price.
  • will save this information in a Django Model (PostgreSQL), to make it easy to retrieve, and don’t re-do work.