Figure 20From: Enriching feature engineering for short text samples by language time series analysisExploratory analysis of features with non-trivial distributions. (a) Log-transformed feature TLD__change_quantiles__f_agg_“mean”__isabs_False__qh_0.6__ql_0.2 extracted from token length distributions. (b) Log-transformed feature TLS__cwt_coefficients__widths_(2, 5, 10, 20)__coeff_14__w_20 extracted from token length sequencesBack to article page