Diabetes Mellitus Prediction with Classification Algorithms Using Weka
Gyawali, Bishal (2024)
2024
All rights reserved. This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:amk-2024060320072
https://urn.fi/URN:NBN:fi:amk-2024060320072
Tiivistelmä
Early diabetes identification is crucial for controlling chronic illnesses. This study uses WEKA to compare the performance of four classification algorithms (Multilayer Perceptron, Logistic Regression, Random Forest, and Extra Trees) for diabetes prediction.
Accuracy, precision, recall, and f-measure were evaluated across various train-test splits. The multilayer perceptron regularly outperformed others, indicating its usefulness in diabetes prediction. Logistic regression and random forest both produced encouraging results. Extra trees have regularly underperformed.
These findings emphasize the potential of classification algorithms for early diabetes diagnosis, which can help healthcare practitioners make more informed decisions. Future research might investigate sophisticated algorithms, combine many data sources, and assess therapeutic impact in real- world scenarios.
Accuracy, precision, recall, and f-measure were evaluated across various train-test splits. The multilayer perceptron regularly outperformed others, indicating its usefulness in diabetes prediction. Logistic regression and random forest both produced encouraging results. Extra trees have regularly underperformed.
These findings emphasize the potential of classification algorithms for early diabetes diagnosis, which can help healthcare practitioners make more informed decisions. Future research might investigate sophisticated algorithms, combine many data sources, and assess therapeutic impact in real- world scenarios.