A Rule-Based Approach for Converting Wikipedia Content into Semantic Relations
NİHAL YAĞMUR AYDIN · Journal of Computers · 2018
In this paper, we propose a method for conversion from natural language into semantic relations.In this research, we focus on text written in highly unstructured form.The tools Stanford named entity recognizer and dependency parser are used to extract relevant information.The method is based on analysis of grammatical patterns of the sentences chosen from Wikipedia.The parse trees of sentences are examined in order to create patterns.Regular expressions are used to fetch the related nodes of the parse trees.In addition to the grammatical structure of sentences, we also made use of the named entities to create semantic relations.Experiments on different types of relations showed that 71% and 82% success rates can be obtained for a threshold of 0.50 correctness rate.