Figure 4From: Enriching feature engineering for short text samples by language time series analysisDiscrimination maps for the first three principle components derived from statistically significant language time series features. The principle components were calculated from all 19,579 sentences. The figures also indicate the location of two example sentences: Sample id13843 (white cross symbol) was written by MWS (cf. p. 13). Sample id15671 (white plus symbol) was written by HPL (cf. p. 19). Both symbols are located in the same bin in panel (c)Back to article page