Answers are grounded in this paper and cite their sources.
This work evaluates AI models for classifying chemical reactions, indicating improved performance and transparency.
The classification of chemical reactions into organic and inorganic types is an important task in cheminformatics and chemical education. In this work, we introduce a new dataset of chemical reactions and evaluate both traditional machine learning models and transformer-based models for classification. Random Forest, SVM, and Logistic Regression achieved moderate performance, while ChemBERTa and RoBERTa provided stronger results. Among these, RoBERTa achieved the highest accuracy of 90%, which is better than ChemBERTa and traditional models. To enhance transparency, we applied Local Interpretable Model-Agnostic Explanations (LIME) for explainable AI, which highlighted reaction components influencing predictions. This contribution provides both high-performing models and interpretable outputs, making it a valuable step toward intelligent chemical reaction classification systems. Future work will focus on extending the dataset and generating chemical structure visualizations fordeeper understanding.
No takes yet. Share an insight, caveat, or question.
Md. Ismiel Hossen Abir (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: