Please use this identifier to cite or link to this item: https://gnanaganga.inflibnet.ac.in:8443/jspui/handle/123456789/15394
Full metadata record
DC FieldValueLanguage
dc.contributor.authorJanarthanan, S-
dc.contributor.authorVarun, T G-
dc.contributor.authorHussain, Shariff Y-
dc.contributor.authorSenbagavalli, M-
dc.date.accessioned2024-04-20T10:53:12Z-
dc.date.available2024-04-20T10:53:12Z-
dc.date.issued2023-
dc.identifier.urihttp://gnanaganga.inflibnet.ac.in:8080/jspui/handle/123456789/15394-
dc.description.abstractMillions of individuals throughout the world suffer with diabetes, a chronic metabolic illness. For diabetes to be effectively managed and complications from developing, early identification and accurate diabetes prediction are essential. In this paper, we suggest a machine learning-based strategy for predicting the development of diabetic illness. The goal is to create an accurate model that can gauge a patient's likelihood of having diabetes based on a variety of medical characteristics. Age, BMI, blood pressure, blood sugar levels, and family history are just a few of the clinical and demographic characteristics we gathered from a wide range of people. Preprocessing was done on the dataset to deal with missing values, outliers, and feature scaling. Several machine learning techniques were used, including random forest, logistic regression, and support vector machines (SVM). Each algorithm's performance was tested using k-fold cross-validation after it had been trained on a portion of the dataset. Metrics for accuracy, precision, recall, and F1-score were used to compare the models. Our findings showed that the random forest method, with an average accuracy of 85% over all folds, had the best prediction accuracy. Indicating the model's capacity to reliably identify people at risk of diabetes, the confusion matrix analysis showed a considerable decrease in false negatives. The relevance of the chosen variables, such as BMI and glucose levels, in the prediction and in the diagnosis of diabetes is highlighted. This work contributes to the field of predicting the development of diabetic illness. The suggested method has potential for helping medical practitioners spot people at risk for diabetes early on, enabling prompt interventions and individualized treatment strategies. It is advised that the model undergo additional validation on bigger and more varied datasets to increase generalizability and boost its effectiveness in actual clinical situations.en_US
dc.language.isoenen_US
dc.publisherAlliance College of Engineering and Design, Alliance Universityen_US
dc.subjectChronic Metabolic Illnessen_US
dc.subjectDiabetes Disease Predictionen_US
dc.subjectMachine Learning Algorithmen_US
dc.subjectSupport Vector Machinesen_US
dc.titleDiabetes Disease Prediction Using Machine Learning Algorithmen_US
dc.typeOtheren_US
Appears in Collections:Dissertations - Alliance College of Engineering & Design

Files in This Item:
File SizeFormat 
CSE_G03_2023.pdf
  Restricted Access
2.2 MBAdobe PDFView/Open Request a copy


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.