Table 2 Correspondence table for applications.
Chinese restaurant metaphor | Reddit data set | Gutenberg data set |
|---|---|---|
Categories | Sentiment categories (positive/negative) | Literary genres (western/history) |
Customer entering the restaurant | Comment | Word |
Chosen table | Author of the commented news | Word |
Occupation of an empty table | First comment to a news | First appearance of a word |
Choice of an already occupied table | Comment to a news whose author has already received comments to other her news | Usage of a word already used before in some of the literary genres |
\(D^*_{t,h}=\) number, until time-step t, of distinct tables occupied for their first time by a customer belonging to category h | Number, until time-step t, of distinct authors whose first received comment belongs to sentiment category h | Number, until time-step t, of distinct words whose first apperance has been in literary genre h |
\(D_{t,h}=\) number, until time-step t, of distinct tables occupied by at least one customer belonging to category h | Number, until time-step t, of distinct authors who have received at least one comment belonging to sentiment category h | Number, until time-step t, of distinct words that have been used in literary genre h |
\(K_t(h,c)=\) number, until time-step t, of customers belonging to category h and sitting at table c | Number, until time-step t of comments received by author c belonging to sentiment category h | Number, until time step t, of times the word c has been used in literary genre h |
