Sentiment Analysis Models with Bayesian Approach: A Bike Preference Application in Metropolitan Cities

Table 1

Sentiment probabilities and observed frequencies evaluation in the numerical example.

Sentiment	Observed frequency (%)		Priori probability (%)		Posterior probability (%)
					With probability distribution of b		Whit expected value of b
		N		N		N		N
User Z	36,4	36,4	79,5	9,7	39,3	37,6	39,7	37,8
User W	46,2	30,8	37,5	0,6	40,1	29,0	41,0	28,4
Average	41,3	33,6	58,5	5,1	39,7	33,3	40,4	33,1