**6.2 Data mining**

There are now a large number of diseases- and drugs-linked information such as gene sequences, protein–protein interactions, and drug–protein interactions with increasing rapid growth, which needs effective approaches to quick access and analysis of hidden information. Commonly, text mining is the most applicable method in the majority of data mining–related studies. In the field of computational drug repurposing, text mining has been used to find the gene, drug, and diseases-related data and then categorize the relevant entities. Regarding drug repurposing, text mining has successfully been used in several studies [40, 41]. Brown et al. suggested an online text-mining server with the ability to drug clustering based on the similarity of their physicochemical properties [42]. A text mining-based tool was also introduced by Leaman et al. for identifying disease-related information mentioned in the literature [43]. In another study, Papanikolaou et al. used text mining to recognize biological entities in the Drug Bank database. The retrieved data were then clustered by different algorithms and used for obtaining novel drug–drug relations [44].
