| Optimised features (number) | Label | Optimised feature |
| Lexical dispersion rates (4) | D4 | DiSp8:0.7–0.8, DiSp9:0.8–0.9, DiSp10 : 0.9–1.0, DiWr10 : 0.9–1.0 | Semantic features (10) | U10 | A2 (words depicting change), A3 (words depicting being/existing), A4 (classification), A6 (comparing), A7 (probability), A13 (degree adverbs), E5 (trepidation, courage, surprise), O4 (physical attributes), Z5 (functional words), Z6 (negative particles). | Sentiment features (10) | L10 | Clout expressions, emotional tones, words per sentences, they (third person pronouns), affect words (incl. positive and negative emotions, anxiety, anger, sad), negative emotions, anxiety, tentativeness, differentiation, core drives and needs (reward focus) | Structural features (5) | S5 | Number of difficult sentences (more than 22 words), number of monosyllabic words, number of long (6+ characters) words, number of sentences which use the same word multiple times (overused words), passive voice | All features (33) | All33 | Dispersion rates (1): DiSp8:0.7–0.8 | Semantic features (21): A3 (being/existing), A13 (degree adverbs), A15 (abstract terms of safety, danger), B2 (physical conditions), B3 (medical treatment), E5 (trepidation, courage, surprise), E6 (apprehension, confidence), K5 (leisure, activities), M1 (movement), M3 (transport on land), M5 (transport by air), M6 (points of reference), N3 (measurement), O4 (physical attributes), S1 (social action, state, process), S5 (affiliation), T1 (time), X2 (reasoning, belief, scepticism), Z3 (organisations names), Z5 (functional words), Z8 (pronouns) | Sentiment features (L4): clout expressions, affect words, negative emotions, anxiety | Structural features (S7): number of sentences using the same words multiple times (overused words), average number of sentences per paragraph, number of questions, number of proper nouns, number of unique multiple (3+) syllable words, number of unique long (more than 6 letters) words, out-of-dictionary words. |
|
|