Enhancing Biomedical Literature Analysis to Predict Drug Interactions Using Multi Criteria Decision Making and LSTM-CNN Model

The extraction of information from unstructured text has become a crucial task in the biomedical literature field. Specifically, the analysis of various relationships such as Drug Food, Drug Disease, Drug non-prescription and Drug Drug interactions has gained significant importance. To accomplish this, a combination of Multi Criteria Decision Making (MCDM) and machine learning techniques has been employed. A dataset has been curated and utilized to train a model based on Long Short-Term Memory (LSTM) and Convolutional Neural Networks (CNN). The trained model enables accurate predictions and identification of relevant information pertaining to drug interactions in biomedical literature. This approach enhances the understanding and knowledge extraction from vast amounts of unstructured text data, leading to improved insights and potential advancements in the field of biomedical research.