Text Sentiment Analysis Based on a New Hybrid Network Model

Table 3

Parameter settings of doc2vec model.

Name	Parameter value	Meaning
dm	1	Training algorithm
Min_count	1	Truncate dictionary
Window	10	The maximum distance between the current word and the prediction word in a sentence
Vector_size	300	Dimension of eigenvector
Sample	0.01	Configuration threshold of random downsampling of high frequency words
Negative	5	Set interference words
Workers	7	Parallel number of parameter control training