Read Abstracts and download

Article Title: An Online Cortical Machine Learning Artificial Intelligence Technique for Drug Discovery


Abstract Bioinformatics deals with the analysis and interpretation of biological data by using tools of information science. Drug discovery prediction which is a process of discovering new candidate medications from some molecular compounds has challenged professionals in the field of medical sciences. Tools that have assisted in drug discovery and have been reported by researchers includes the use of decision trees, induction programming logic, expert systems and supervised neural networks. In this research paper, we propose an approach to the drug discovery and prediction problem using a variant of an unsupervised online cortical machine learning artificial intelligence technique. The approach has an explicit tuning parameter called the relaxation factor used in determining possible new candidate sequence. Experiments on a popular DNA sequence dataset and a reversed-phase high-performance liquid chromatography (RP-HPLC) drug dataset were performed to determine whether the proposed technique can give effective predictions. The results showed that the approach compares favorably with the other methods reported in the literature but has a more promising performance when it is set to lower relaxation values