Research on Sentiment Tendency and Evolution of Public Opinions in Social Networks of Smart City

Table 2

Experiment datasets.

The dataset (the number of comments)	Number of words in per microblog		Vocabulary size
	Initial	Pretreatment	Initial	Pretreatment
The dataset 1 (3562)	134	102	9789	6319
The dataset 2 (3527)	127	94	9736	6242
The dataset 3 (3617)	131	100	9780	6301
The dataset 4 (3582)	128	96	9742	6254
Average	130	98	9762	6279