Table 13. Benchmark of different ensemble learning strategies with the whole corpus evaluated for sentiment towards the society.

Weighted precision (W-P), weighted recall (W-R), weighted (W-F1) and macro (M-F1) are reported.

	W-P	W-R	W-F1	M-F1
Max	0.7246	0.7236	0.7231	0.7221
Mean	0.7373	0.7366	0.7366	0.7375
Mode	0.7319	0.7314	0.7315	0.7330