Downloads: 0
Original Research | Biochemistry | Volume 15 Issue 6, June 2026 | Pages: 671 - 673 | India
Evaluation of Biomarkers Dataset for Patients Consumed With and Without Antioxidant: An Approach by Machine Learning Algorithms
Abstract: Considerable attention has been paid to assessing the impact of antioxidants on different biological markers due to their increasing application in the prevention and treatment of illnesses brought by oxidative stress. Complex, nonlinear correlations between several biomarkers cannot be identified using conventional statistical techniques. Machine learning (ML) algorithms offer advanced analytical techniques for recognizing latent patterns, categorizing patient groups, and forecasting treatment outcomes. This research seeks to use ML techniques to compare biomarker data from individuals who took antioxidants with those who did not. In the present study, biomarker dataset mining through ML algorithms viz. Bayes Network (BN), NaiveBayes (NB), Logistic Regression (LR), K-nearest neighbour (IBK), Instance-based classifier (K*) and LogitBoost (LB) were studied by using WEKA (Waikato Environment for Knowledge Analysis) tool (version, 3.8.5). The prediction of accuracy as per effect class (normal and abnormal) values related to statistical interpretations on MCC, ROC and PRC in which the highest values were predicted in K* and LB followed by BN, NB, IBK and LR. By improving knowledge of antioxidant effectiveness and promoting clinical decision-making, the findings have the potential to further personalized medicine.
Keywords: Algorithms, Biomarkers, Dataset, Machine learning, Prediction accuracy
How to Cite?: Sk Ziaur Rahaman, "Evaluation of Biomarkers Dataset for Patients Consumed With and Without Antioxidant: An Approach by Machine Learning Algorithms", Volume 15 Issue 6, June 2026, International Journal of Science and Research (IJSR), Pages: 671-673, https://www.ijsr.net/getabstract.php?paperid=SR26611163636, DOI: https://dx.dx.doi.org/10.21275/SR26611163636