Implementation of Random Forest Algorithm to Determine Food Allergies

Ratu Aisyah
Dede Brahma Arianto

Ambidextrous : Journal of Innovation, Efficiency and Technology in Organization
Vol. 4 No. 02 (2026): Ambidextrous: Journal of Innovation, Efficiency and Technology in Organization

Food allergy diagnosis faces challenges due to symptom variation and delays in conventional testing. This study aims to classify food allergy types using the Random Forest algorithm on patient data including age, gender, food type, symptoms, and severity. A quantitative experimental design was implemented with a secondary dataset of 1,000 medical records as the population, divided through stratified sampling (train-test ratio 80:20). Data preprocessing included label encoding of categorical variables, followed by supervised classification analysis using Python scikit-learn. The results showed a model accuracy of 85%, precision of 84%, recall of 86%, and F1-score of 85%, indicating strong performance with a balanced error rate in the validation confusion matrix. In conclusion, Random Forest effectively supports the rapid identification of food allergies, potentially serving as a clinical decision-making tool with the need for larger prospective datasets.