**6.3 Machine learning (ML)**

Machine learning, a crucial subset of artificial intelligence (AI), has been combined into many fields, such as data generation and analytics. Related to drug discovery, ML algorithms may participate in target and lead discovery as well as develop quantitative structure–activity relationships. Briefly, in machine learningbased drug repositioning, different algorithms, such as artificial neural networks (ANNs), support vector machines (SVMs), and random forest (RF), were trained by numerical forms of different features of drugs, diseases, genes, and so on. The trained algorithms can then predict the drug ability of unknown compounds [45]. In this regard, Gottlieb et al. used drug–drug and disease–disease similarity events as grouping features for training a logistic regression classifier and prediction of drug-disease associations [46]. Similarly, Napolitano et al. introduced a SVM model trained by drug-related similarities with the ability to forecast the therapeutic class of United States Food and Drug Administration (FDA)-approved compounds [47]. Aliper et al. introduced a fully connected deep neural network algorithm trained

*Evaluation of Drug Repositioning by Molecular Docking of Pharmaceutical Resources… DOI: http://dx.doi.org/10.5772/intechopen.101395*

by gene expression signatures for predicting therapeutic potentials and new drug suggestions [48].
