Using morphological features to simplify complex sentences in Punjabi language

Abstract

In this research paper, a technique for automatic identification and simplification of long and complex sentences in Punjabi language has been proposed. In this research, mainly compound and complex sentences have been simplified. Various morphological features of non-finite verb and type of conjunctions used in complex sentences are used for identification of complex sentences. Sentence structure and conjunction are used for identification of compound sentences. Structure of clauses has been used to mark the start and end position of the dependent and independent clauses present in complex sentences. Syntactic technique has been used for simplification of compound and complex sentences. Author tested this system on compound and complex sentences and obtained an accuracy of 98% in case of compound sentences and 78% for complex sentence.