Answers are grounded in this paper and cite their sources.
Paraphrasing normally involves sophisticated external resources and semantic thesauri. This paper describes the automatic generation of Modern Greek paraphrases using statistical significance testing for extracting applicable syntactic reordering schemata. Next, supervised filtering helps remove erroneous schemata minding their context. As the paraphrases will be used in steganographic communication, they need not be sophisticated alterations, but significant in number. The proposed process is therefore knowledge-poor (portable to other languages with similar syntax), robust and domain-independent.
No takes yet. Share an insight, caveat, or question.
Katia Lida Kermanidis (2010) studied this question.
Synapse has enriched one closely related paper. Consider it for comparative context: