Big Data Framework for Predicting Infectious Diseases to improve Healthcare by Discovering New Symptom Patterns

Abstract

Infectious disease control is one of the most thrilling opportunities form using big data, where these streams of novel data can be used to improve timeliness for preventing. Various public and private sector Healthcare providers generate, store, and analyse big data to improve the services they provide. Lately, the COVID-19-new Corona virus outbreak has put human health, life, production, social connections, and international relations in grave danger. Consequently, big data technologies have been crucial in the pandemic response. Infectious disease occurs when a person has a disease by a pathogen from another person. It is a problem that causes harm for both individual and macro scales. In addition, infectious illness patterns are unknown, which complicate the prediction process. This study aims to create big data framework to predict infectious diseases by discovering new symptoms patterns to enhance healthcare infection prevention and control. To achieve this goal, machine learning algorithms K-Nearest Neighbors (K-NN) and Random Forest (RF) were used to clean and maintain big data from December 2019 to June 2020. Additionally, the mining model FP-growth and Park, Chen, and Yu (PCY) of China were applied to discover new symptom rules. The results show that the RF model performs better than K-NN with accuracy rates of 97%, and the PCY model performs better than FP-growth with an accuracy rate of 98%. These results highlight the potential of big data and machine learning in identifying patterns and predicting infectious diseases, which can ultimately improve public health outcomes.