Research Article

A Supervised Approach to Predict the Hierarchical Structure of Conversation Threads for Comments

Table 2

Datasets’ statistics split up per website.

DatasetsAvg. length of comments’ text (in words)Avg. length of the root’s text (in words)Number of commentsAvg. comments per newsNumber of newsNumber of commenter

Thestandard72.98357.6814905849.8129923431
Alef74.66766.552321027.968304387
ENENews68.19162.641666138.92428831
Courantblogs79.24398.17495219.192581716
Russianblog18.56237.163087716.1819081258