Data Mining Big Data Inpatient Database Using Cuckoo Search

Abstract

This paper investigates data mining in a medical dataset by using the stochastic population-based nature-inspired Cuckoo search algorithm. Particularly, association rules are mined by applying an objective function composed of support and confidence weighted by two parameters for controlling the importance of each measure. The rules are mined in a Nationwide Inpatient Sample dataset, which is a collection of discharge records of several hospitals in the USA. Only those records, where a patient was diagnosed with Type II diabetes mellitus were extracted for association rule mining. The results show that the found rules are simple, easy to understand and also interesting, as they were verified with actual clinical studies. The results obtained can be beneficial to either doctors or insurance companies.