• Mr. Paluru Chakali Pavan Kumar
• Mr. Dr R. Kaviarasan

An relationship between two medications in which one drug's pharmacological actions are altered by another is known as a drug-drug interaction (DDI). Positive DDIs typically enhance patients' therapeutic outcomes, however negative DDIs are the primary culprits behind unfavorable medication responses, which may possibly lead to the drug's recall and the patient's demise. As a result, DDI identification has become crucial to the development of new drugs and the management of disease. In this paper, we offer an innovative technique called DDI-IS-SL, which is based on integrated similarity and semi-supervised learning. The cosine similarity method is used by DDI-IS-SL to combine the drug's chemical, biological, and phenotypic data in order to determine how similar the medications' features are. Additionally, the kernel similarity of medicines' Gaussian Interaction Profiles is determined using known DDIs. The Regularized Least Squares classifier is used to determine the interaction probability scores of drug-drug pairings using a semi-supervised learning approach. By way of In comparison to previous approaches, the DDI-IS-SL can obtain better prediction performance using the 5-fold, 10-fold, and de novo drug validation. Additionally, DDI-IS-SL takes less time on average to compute than other comparable approaches. Finally, case studies provide more evidence of how well DDI-IS-SL performs in real-world settings.

Machine Learning Based Drug Interactions Prediction

"Machine Learning Based Drug Interactions Prediction", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.10, Issue 7, page no.g670-g673, July-2023, Available :http://www.jetir.org/papers/JETIR2307689.pdf

"Machine Learning Based Drug Interactions Prediction", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.10, Issue 7, page no. ppg670-g673, July-2023, Available at : http://www.jetir.org/papers/JETIR2307689.pdf