Application of formal grammar in text mining and construction of an ontology

This work describes an investigation of formal grammar with application to text mining. It is an important area since text is the most widespread type of data and it contains a lot of potentially useful information. Unstructured nature of text requires other methods for its processing, in contrast to other types of data mining. In this work, the authors propose an original approach to text mining by making a parse tree for each sentence using regular grammar and creating an ontology and provide a demonstration of this system being implemented in a constrained scenario. This ontology can be used for different tasks, ranging from expert systems to automatic machine translation. The ontology is a network consisting of concepts linked by relations. The authors developed a new system to implement proposed approach working in different languages.