In my previous post NGrams and Edge NGrams – computational complexity, I explored the runtime implications of working both with NGrams and Edge NGrams. This is a follow-up post that will build on the previous one so if you are not familiar with these terms, I would suggest checking it first. In this post, I would like to take a look at the space requirements of NGrams and Edge NGrams and how to estimate the size required to store the data intended for your indices.Continue reading “NGrams and Edge NGrams – space complexity”
In my current project, we do a bunch of work revolving around Elasticsearch and enabling our customers to quickly access the relevant portions of our large data set. A couple of weeks ago I was asked to come up with a method to compare the costs of working with NGrams and Edge NGrams. I tried to make my life easier and look around the Internet for somebody else’s breakdown but I didn’t find anything I would like. So I decided to bite the bullet and do the work myself. In this post and the follow up one, I would like to present my way of reasoning about NGrams and Edge NGrams.