communities (2).pickle -> contains the final clustering

NER - Story Level Dataset - Single Paper -> use communities (2).pickle to create a story chain level dataset, e.g. number of arts, duration

NER - Article Level Dataset - Paper Level Focus -> use communities (2).pickle to create an article level dataset, e.g. how many articles are in a story chain

Both of the abpve have a 'paper level' focus -> i.e. only counting story chains within the same paper

NER - Analysis - CMM Final May 2018.r -> take the datasets from the above two files and produce analytics for the paper 

Clusters - IRR.r -> analyse IRR scores for the 'intra-cluster' check