An Improved Apriori Algorithm Applied to Mining Ancient Chinese Poems

摘要

Based on the study of classical Apriori algorithm, this paper makes a comparison between FP-tree algorithm and Eclat algorithm, analyzes the basic principles of the three algorithms and their respective advantages and disadvantages. After the research on the algorithms, the author finds that time and space optimization can not have both. So the author puts forward an improved algorithm with the conception of importance inspired by the rules of Chinese poems, and puts forward the concept of “importance degree” with examples. Compared to the classic Apriori algorithm, the new improved algorithm ignores some association rules that are not concerned by the user, and greatly reduces the generation of two candidate items. The running speed of the program is increased by up to hundreds of times in the testing process. The improved algorithm can be applied to explore the association rules of the characters in Chinese poems, which is of great significance to the study of ancient poetry and grammar.