DSpace Repository

Diabetes Prediction Using Data Mining

Show simple item record

dc.contributor.author Khanom, Khadiza
dc.contributor.author Mou, Farzana Hasnat
dc.contributor.author Mim, Firuj Samiha
dc.date.accessioned 2021-09-01T07:08:30Z
dc.date.available 2021-09-01T07:08:30Z
dc.date.issued 2020-12-31
dc.identifier.uri http://dspace.daffodilvarsity.edu.bd:8080/handle/123456789/6060
dc.description.abstract This current epoch is facing a hard nut to crack which is none other than diabetes. Increasing blood sugar is the main issue of this deadliest disease. We get glucose from our food what we eat. Insulin is a hormone. It helps to get sugar or glucose from blood into our cell and produce energy. If there are lack of insulin and glucose cannot be got in cell then it remains left over in blood and that’s why the glucose level in blood increased. This is mainly known as diabetes. Diabetes is dependent on our lifestyle. When we are not aware of our diet, weight and body fitness, we use to consume extra sugar through our food and the extra sugar cannot be utilized to produce energy and they remain same that causes glucose high. Proper production and management of Insulin hormone is necessary. There are two form of diabetes such as type 1 and type 2. Diabetes causes so many complication in our body. It affect our kidney, eyes, heart and so on. It is good to be aware of having diabetes as well as lead a controlled life. When it is too late to identify diabetes, it causes so many difficulties. Consulting a doctor and visiting diagnostic centers is a tiresome process for a patient. But now-a-days the improvement of data mining and machine learning approaches find out a solution of this troublesome issue. This paper is the description of our study to predict diabetes of women in an early stage using data mining algorithms. To make us work, we have used six data mining algorithms namely K Nearest Neighbors, Decision Tree, Logistic Regression, Support Vector Classifier, Naïve Bayes and Random Forest. We have experimented on Pima Indians Diabetes Database (PIDD) that is chosen from Kaggle which contains 768 record and 9 features (first 8 are independent and the last one is dependent), a community of data science. K Nearest Neighbors algorithm are in the utmost position considering accuracy. Its accuracy rate is about 78.57% which is a far cry from the other algorithms. This exploration are acting by some ordinary factors accountable for this chronic disease as for instance pregnancies, glucose, blood pressure, BMI, insulin, age, etc. en_US
dc.language.iso en_US en_US
dc.publisher Daffodil International University en_US
dc.subject Diabetes--Treatment en_US
dc.subject Data mining en_US
dc.title Diabetes Prediction Using Data Mining en_US
dc.type Other en_US

Files in this item

This item appears in the following Collection(s)

Show simple item record

Search DSpace


My Account