This function extracts the dates for which aggregated time series sentiment is most extreme (lowest, highest or both in absolute terms). The extracted dates are unique, even when, for example, all most extreme sentiment values (for different sentiment measures) occur on only one date.
peakdates(sento_measures, n = 10, type = "both", do.average = FALSE)
sento_measures | a |
---|---|
n | a positive |
type | a |
do.average | a |
A vector of type "Date"
corresponding to the n
extracted sentiment peak dates.
Samuel Borms
set.seed(505) data("usnews", package = "sentometrics") data("list_lexicons", package = "sentometrics") data("list_valence_shifters", package = "sentometrics") # construct a sento_measures object to start with corpus <- sento_corpus(corpusdf = usnews) corpusSample <- quanteda::corpus_sample(corpus, size = 500) l <- sento_lexicons(list_lexicons[c("LM_en", "HENRY_en")], list_valence_shifters[["en"]]) ctr <- ctr_agg(howTime = c("equal_weight", "linear"), by = "month", lag = 3) sento_measures <- sento_measures(corpusSample, l, ctr) # extract the peaks peaksAbs <- peakdates(sento_measures, n = 5) peaksAbsQuantile <- peakdates(sento_measures, n = 0.50) peaksPos <- peakdates(sento_measures, n = 5, type = "pos") peaksNeg <- peakdates(sento_measures, n = 5, type = "neg")