Judging Subjective and Objective Sentence Based on 2-POS Subjective Models

With the constant increasing of the TV programs text evaluate data on Internet, using emotional tendentiousness analysis technology to analyze the mass text data will help the program producers, program promulgator, network operator and government department improve their competitiveness and make a better decision, accordingly improve the competitiveness of the broadcasting and TV. The paper is on the basis of part of speech combination model. We firstly use the Chi-square statistic method to extract 2-POS subjective models, then calculate sentence´s subjective degree, finally judge the sentence as subjective or objective sentence by comparing the degree with the threshold.