Objective The factors influencing the frequency of acute exacerbation of chronic obstructive pulmonary disease (COPD) were studied by principal component analysis (PCA) and discriminant analysis in order to predict the frequency of acute exacerbation in patients with chronic obstructive pulmonary disease (COPD). Methods 350 patients with chronic obstructive pulmonary disease (COPD) admitted in our hospital from January 2016 to October 2017 were collected. Immediate general data including age, height, weight, medical history, smoking year and arterial blood gas analysis were collected. For the first time, content of procalcitonin (PCT) and the content of Creactive protein (CRP) and blood routine were collected. The antibiotic treatment time and hospitalization time at the end of admission were collected. The patients were followed up for one year and the frequency of acute exacerbations of chronic obstructive pulmonary disease (AECOPD) was determined. According to the frequency of exacerbations, the patients were divided into group A (≤ one time), group B (＜three times＝ and (group C ≥three times). The differences of general data and clinical data among the three groups after admission were compared. Clinical indexes and general data were extracted by factor analysis. The frequency of acute exacerbation of AECOPD patients was analyzed by discriminant analysis using of above data and the extracted principal components. The difference of discriminant analysis rate before and after the extraction of principal components was compared. Results The general data and clinical data of the three groups had no statistical difference except smoking years, but there was statistical difference in the other indexes among the three groups (P＜0.05). Five principal components were extracted from clinical indexes and general data, and the extraction rate of comprehensiveinformationwas 83.3%. The general data and clinical index discriminant analysis showed that the overall correct discriminant rate was 74.8%, and the correct discrimination rate of the three groups was 88.6%,65.7% and 70.3%. The discriminantanalysis using the extracted principal components suggested that the correct rate of overall discrimination was 72%, and the correct discrimination rates of the 3 groups were 97.1%, 60%,and 59.5%. Conclusion The patients with chronic obstructive pulmonary disease have different characteristics according to the different frequency of acute exacerbation. The principal representative components can be extracted by factor analysis from common clinical collected indexes. There is no significant difference between the discriminant accuracy rate of principal component analysis and the comprehensive data.