master
Erik de Vries 4 years ago
parent 09fd8d0cb2
commit f6dfc6711b

@ -208,7 +208,7 @@ sent_merger <- function(df, actors_meta = NULL, actor_groups = NULL, pos_cutoff
parties <- actors_meta[parties, on = c('ids'), mult = 'first'][!is.na(id),.(ids = str_c(ids,"_mfs"), (.SD)), .SDcols = -c('ids')] parties <- actors_meta[parties, on = c('ids'), mult = 'first'][!is.na(id),.(ids = str_c(ids,"_mfs"), (.SD)), .SDcols = -c('ids')]
## Join all aggregations into a single data frame, compute derived actor-level measures, and add date dummies ## Join all aggregations into a single data frame, compute derived actor-level measures, and add date dummies
df <- bind_rows(actors, parties, parties_actors, all, all_ind, all_par) %>% df <- bind_rows(actors, parties, parties_actors, all) %>%
left_join(.,text_sent, by=c("id","publication_date")) %>% left_join(.,text_sent, by=c("id","publication_date")) %>%
left_join(.,text_noactors, by="id") %>% left_join(.,text_noactors, by="id") %>%
mutate( mutate(

Loading…
Cancel
Save