Commonsense knowledge extraction for Persian language: A combinatory approach

Putting human commonsense knowledge into computers has always been a long standing dream of artificial intelligence (AI). The cost of several tens of millions of dollars and times have been covered so that the computers could know about “objects falling, not rising.”,” running is faster than walking. The large database was built, automated and semi-automated methods were introduced and volunteers’ efforts were utilized to achieve this, but an automated, high-throughput and low-noise method for commonsense collection still remains as the holy grail of AI. The aim of this study was to build commonsense knowledge ontology using three approaches namely Hearst method, machine translation and using structured resources. Using three Persian corpuse and Applying aforementioned methods, we could extract 7 different relations. 70000 assertions have been extracted. Finally, average accuracy of Hearst, MT and structured resource were 75%, 75% and 100% respectively.