Dependency parsing based Chinese open relation extraction
Jing Xu, Liang Gan, Lu Deng, Jing Wang, Yan Zhou · 2015
The traditional relation extraction methods require the pre-defined relation types and a corpus with human tags. The information extracted by the current open relation extraction (ORE) methods is incomplete, and the relation types are finite. To solve the above problems, we propose ClausORE, which is an n-ary ORE method for Chinese text and extracts the entities and relations between entities from natural language text. ClausORE exploits Chinese syntactic features to formulate the extraction rules and deal with the complex sentences especially, which have complex structures and contain two or more clauses. The relation tuples extracted by ClausORE ensure the information integrity and avoid the specificity. The experimental results over the data set of real news indicate that ClausORE is an ORE method with high precision, which could capture more information than the existing ORE methods.