Contextual Biomedical Language Models for Imbalance-Aware Drug-Food Interaction Classification
- 1Neil Gogte Institute of Technology
Abstract
Drug-food interactions (DFIs) are essential to patient safety because they can alter drug effectiveness or cause side effects. Automated classification of DFIs is difficult because of unstructured textual data, semantic ambiguity, and class imbalance, especially when separating neutral interactions. This paper presents an imbalance-aware deep learning approach for classifying drug-food interactions into three classes: safe, neutral, and unsafe. A structured dataset is created from textual drug interaction descriptions by extracting drug-food pairs and labeling them using rule-based heuristics. To address class imbalance, we apply class weighting and focal loss while fine-tuning the context-aware BioBERT language model. Our experiments show that the approach reaches an accuracy of 85% and a macro-averaged F1-score of 0.77, with a notable improvement on the minority neutral class (F1-score of 0.56). The main novelty of this work is the combination of contextual biomedical language modeling with imbalance-aware deep learning, enabling more effective classification of DFI datasets.
Related
- Tracking claim changes from preprint to publication across 72,644 biomedical studies using large language models
- Comparing Temporal Averaging and Transformer Encoders for Video-Based Cattle Behavior Classification Under Severe Class Imbalance.
- A Comparative Benchmark of Biomedical Language Models for Concept Normalization from Real-World Text
- Quantifying performance inflation from hidden sample dependence in biomedical image classification benchmarks
- Application of Bayesian-Derived Gene Interaction Network and Subnetwork for Classification of Idiopathic Parkinson Disease with Biomarkers
BibTeX
@misc{nalgonda2026contextual,
title = {{Contextual Biomedical Language Models for Imbalance-Aware Drug-Food Interaction Classification}},
author = {L. Nalgonda and M. Yadav and Z. Mohammed and M. Abdullah},
howpublished = {bioRxiv},
year = {2026},
month = oct,
doi = {10.64898/2026.10.19.760076},
url = {https://www.biorxiv.org/content/10.64898/2026.10.19.760076v1},
note = {CC BY-NC-ND},
}
RIS
TY - UNPB
TI - Contextual Biomedical Language Models for Imbalance-Aware Drug-Food Interaction Classification
AU - L. Nalgonda
AU - M. Yadav
AU - Z. Mohammed
AU - M. Abdullah
PY - 2026
DA - 2026/10/19
PB - bioRxiv
DO - 10.64898/2026.10.19.760076
UR - https://www.biorxiv.org/content/10.64898/2026.10.19.760076v1
AB - Drug-food interactions (DFIs) are essential to patient safety because they can alter drug effectiveness or cause side effects. Automated classification of DFIs is difficult because of unstructured textual data, semantic ambiguity, and class imbalance, especially when separating neutral interactions. This paper presents an imbalance-aware deep learning approach for classifying drug-food interactions into three classes: safe, neutral, and unsafe. A structured dataset is created from textual drug interaction descriptions by extracting drug-food pairs and labeling them using rule-based heuristics. To address class imbalance, we apply class weighting and focal loss while fine-tuning the context-aware BioBERT language model. Our experiments show that the approach reaches an accuracy of 85% and a macro-averaged F1-score of 0.77, with a notable improvement on the minority neutral class (F1-score of 0.56). The main novelty of this work is the combination of contextual biomedical language modeling with imbalance-aware deep learning, enabling more effective classification of DFI datasets.
ER -