Big Data Aspect-Based Opinion Mining Using the SLDA and HME-LDA Models

Table 1

The description of related symbols in the SLDA model.

Symbol	Description
	The total number of comments in the corpus. The unit of corpus is the sentence
	The number of topics
	The number of words in the corpus
	The -th word in the -th comment in the corpus
	The topic of -th comment. The value is
	An indicator variable. The value is . It is used to indicate the opinion target words and the sentiment opinion words.
	An indicator variable. The value is . It is used to indicate positive and negative sentiment opinion words.
	The opinion target, the sentiment opinion word, the positive sentiment word, the negative sentiment word
	The distribution of the opinion target generated by a priori Dirichlet distribution with the parameter
	The distribution of positive comment words generated by a priori Dirichlet distribution with parameters
	The distribution of negative comment words generated by a priori Dirichlet distribution with parameters
	The distribution of sentence topic terms generated by a priori Dirichlet distribution with parameter