Extracting Chemical Laws from Chinese Scientific Literatures

The knowledge acquisition bottleneck has become the major impediment to the development and application of effective knowledge systems. In this paper, we study the problem of extracting chemical laws from the abstract of Chinese scientific and formalizing the laws extracted to production rules. This task is different from traditional text based information extraction because how to encode the extracted laws into a computational chemistry expert system is also under our consideration. NLP approaches are used to facilitate the automated extraction of chemical laws. First, tregex are used to identify the named entities on parsing trees and then typed dependencies are utilized to extract operation words and consequence words. The experimental results indicate that our approaches for this task are highly efficient.