Developing a computer vision system for automated medication identification and verification

Drug repurposing is the process by which novel medications are discovered to cure or prevent illness. Furthermore, biological research has benefited from the astounding advancements in medicine brought about by new and sophisticated technologies. In order to better utilize the vast quantity of biological data that is currently available and to reduce the time and cost associated with the drug discovery and development process, numerous computational techniques are being used. As a result, attempts are being made to carry out drug repurposing, which comprises developing new uses for previously approved medications. To find new approaches for this kind of innovation, a number of computer techniques have been developed to forecast drug-target interactions (Drug target Interactions). Because the data sets gathered from Drug and Targets were so complex, traditional machine learning techniques like linear regression analysis were unable to analyze them effectively. Presenting several machine learning categorization models is the aim of this thesis. This thesis's experiments and assessments show the development of an integrative machine learning model for predicting and finding novel biomedical relationships from both text datasets.