An Apriori-Algorithm-Based Analysis Method on Physical Fitness Test Data for College Students

EasyChair Preprint 4522

Abstract

Since being required to carry out physical fitness tests for students, colleges and universities have accumulated a huge amount of data to deal with annually. However, it is almost impossible to discover the potential relationship among the indicators in physical fitness tests by adopting traditional data processing and analysis methods. Hence, how to identify potential information from the test data and seek corresponding solutions has become the key to improving students’ physical fitness and teaching quality. The Apriori algorithm as a classic approach explores the relationship among a huge amount of data. For the purpose of improving computational performance in addressing a large number of redundant candidate sets through the use of the traditional Apriori algorithm, a method is accordingly developed in this paper. It first determines the attribute significance in test data by using decision tree classifier. Then, it deletes the attributes with lower significance in the test data, reduces the number of scans of data sets, decreases the number of candidate sets, and generates the corresponding association rule. Finally, an experiments conducted to verify the effectiveness of the proposed algorithm.

Keyphrases: Apriori algorithm, Physical Fitness Test, association rule, attribute significance

@booklet{EasyChair:4522,
  author    = {Shupo Nan and Maojian Chen},
  title     = {An Apriori-Algorithm-Based Analysis Method on Physical Fitness Test Data for College Students},
  howpublished = {EasyChair Preprint 4522},
  year      = {EasyChair, 2020}}