Mining Healthcare Data to Predict Diabetes Incidence: Report

This report explores the application of data mining techniques in healthcare to predict the incidence of diabetes. It highlights the motivation behind this research, driven by the increasing global mortality rate due to diabetes and the availability of large datasets. The report discusses various data mining techniques, such as Naive Bayes, Decision Trees, and neural networks, used for diagnosis and treatment. It addresses key research questions, including the current problems in diabetes diagnosis and treatment, gaps in research, the reliability of mining techniques, and the best aspects of diabetes symptom prediction using data mining. The conclusion emphasizes the effectiveness of data mining in extracting valuable knowledge from data, contributing to early disease detection, prevention, and cost savings. The report also references previous research on diabetes diagnosis and treatment, and provides a comprehensive list of references. This study aims to contribute to the development of tools and techniques that can assist healthcare professionals in diagnosing and treating diabetes-related diseases more effectively.

Running head: MINING HEALTH CARE DATA TO PREDICT INCIDENCE OF DIABETES
Mining Health care data to predict incidence of diabetes
Name of Student:
Name of University:
Author’s Note:

1MINING HEALTH CARE DATA TO PREDICT INCIDENCE OF DIABETES
Table of Contents
Introduction......................................................................................................................................2
Motivation........................................................................................................................................2
Conclusion.......................................................................................................................................3
Reference List..................................................................................................................................5

2MINING HEALTH CARE DATA TO PREDICT INCIDENCE OF DIABETES
Introduction
The use of “Data mining” is evident in the examination of large datasets for extracting
concealed and unexplored data patters, knowledge and relationship. This method is conducive in
gathering information where the traditional statistical method is not useful. “Data mining in
healthcare” is an emerging concept which has led to better understanding of the medical data. In
general, the rapid growing success of the application of the data mining tools is observed in wide
range of application such as “analysis of organic compounds, financial forecasting, healthcare
and weather forecasting”. The practice of data mining in “health-care centres” is ensures better
health policy-making and “prevention of hospital errors”. The application of the data mining
concept is further identified with “early detection of the disease, disease prevention, detecting
fraudulent insurance claims, cost savings and ensuring more value for money”. As per the
various types of the previous empirical research evidence, the use of “data mining” techniques
are used for the “diagnosis” of different diseases. Some of the most evident form of the disease
are seen with the diagnosis of “diabetes, stroke, cancer, and heart disease” (Raghupathi 2016).
Motivation
The researchers are seen to be motivated by the worldwide increase in the mortality rate
due to diabetes worldwide in the last 10 years. The increasing mortality rate due to diabetes
every year and huge availability of data has led to extract useful and knowledgeable information
by using data mining techniques for assisting the healthcare specialists in the diagnosis of
diabetes. Some of the other reasons for the motivation is taken into account with the developing
a tool to be embedded in the hospitals management to provide information to the healthcare
professionals (Tsai et al. 2014). This is seen in providing suitable treatment and diagnosis of the
diabetes associated diseases. This is observed with diagnosis of diabetes disease such as “Naïve
Bayes, Decision Tree, neural network, kernel density, automatically defined groups, bagging
algorithm, and support vector machine showing different levels of accuracies” (Chaurasia 2017).
The application of “data mining” in the diagnosis and treatment is helpful for the
identification of the research plans among diabetes patients to formulate a treatment plan. It is
further seen that the hospitals are not seen to provide equal quality in the diabetes related disease.

3MINING HEALTH CARE DATA TO PREDICT INCIDENCE OF DIABETES
Henceforth, it is important that a suitable research is conducted to Diabetes disease professionals
are having sufficient information of the patient’s data. It is also considered to be useful in
analysing the “datasets” to extract the valuable knowledge. The data mining is considered as an
active tool for the analysing the data to extract the useful knowledge. Some of the main form of
the data mining techniques are seen with clustering of the task, maintaining classification trees
and producing rule based algorithms which will be conducive in producing a set of rules
implemented to classify data (Jothi and Husain 2015).
Research Questions
The research aims to identify and address the following research questions:
 What are the main present problems in research techniques in identifying diabetes disease
diagnosis and treatment procedures?
 What are “gaps” in in the “research on diabetes disease diagnosis” and treatment?
 How reliable are mining techniques to diabetes disease treatment in controlling and
diagnosis of diabetes disease?
 Which aspect of the prediction of diabetes symptoms are best traced with the application
of data mining techniques?
Conclusion
It has been discerned that “data mining” is considered as an effective tool for the
analysing the data to extract the useful knowledge. Some of the main form of the data mining
techniques are seen with clustering of the task, maintaining classification trees and producing
rule based algorithms which will be conducive in producing a set of rules implemented to
classify data. The final assessment will be able to contribute to the early detection of the disease,
disease prevention, detecting fraudulent insurance claims, cost savings and ensuring more value
for money. The research will be able to further contribute to the study of “diagnosis of diabetes
disease” such as “Naïve Bayes, Decision Tree, neural network, kernel density, automatically
defined groups, bagging algorithm, and support vector machine showing different levels of
accuracies”. The main research questions will be addressed with the identification of the research
gaps p in identifying diabetes disease diagnosis and treatment procedures. In addition to this, the

4MINING HEALTH CARE DATA TO PREDICT INCIDENCE OF DIABETES
study will be able to state on the reliability of the mining techniques to diabetes disease treatment
in controlling and diagnosis of diabetes disease.

5MINING HEALTH CARE DATA TO PREDICT INCIDENCE OF DIABETES
Reference List
Chaurasia, V., 2017. Early prediction of heart diseases using data mining techniques.
Jothi, N. and Husain, W., 2015. Data mining in healthcare–a review. Procedia Computer
Science, 72, pp.306-313.
Raghupathi, W., 2016. Data mining in healthcare. Healthcare Informatics: Improving Efficiency
through Technology, Analytics, and Management, pp.353-372.
Tsai, C.W., Lai, C.F., Chiang, M.C. and Yang, L.T., 2014. Data mining for Internet of Things: A
survey. IEEE Communications Surveys and Tutorials, 16(1), pp.77-97.