Clinical Significance of Measles and Its Prediction Using Data Mining Techniques: A Systematic Review

Abstract

Data science techniques used in the past era for data extraction are now being replaced by data mining methods due to a lot of contemporary trends and challenges. Data mining includes implicit data extraction, which is previously unknown but potentially useful information, with the help of databases to generate new information. Medical data mining is an area of challenges since the data involves in it is imprecise, inconsistent and massive. Medical diagnostics systems are evaluated by employing large information databases, but it endures many failures to extract data from the databases. There exists no tool to discover the major relationships concerning the data. In such a case, the core knowledge of healthcare data is extracted by applying the data mining methods, thus helping in turning raw data into useful information. Measles is an immunizable disease most likely seen in infants and young children. Measles (rubeola), is an RNA virus, belongs to the Morbillivirus genus, and Paramyxoviridae family. It is one among the six major killer diseases, with accounts to 90% secondary infection rate with susceptible contacts. In this paper, the authors present a systematic review of measles cases and its mortality with the help of the taxonomical tree structure. The published research which utilizes the application of data mining techniques for early prediction of measles in terms of methods used, algorithms utilized and results obtained are evaluated. The paper provides major consolidation of data with respect to motive, study and methods used in the literature. The authors present the summary of the research review findings and research gaps for further study and possible application.