Pattern-Based Acquisition of Scientific Entities from Scholarly Article Titles

We describe a rule-based approach for the automatic acquisition of salient scientific entities from Computational Linguistics (CL) scholarly article titles. Two observations motivated approach: (i) noting aspects an article's contribution in its title; and (ii) pattern regularities capturing terms that could be expressed set rules. Only those lexico-syntactic patterns were selected easily recognizable, occurred frequently, positionally indicated entity type. The rules developed on collection 50,237 CL titles covering all articles ACL Anthology. In total, 19,799 research problems, 18,111 solutions, 20,033 resources, 1,059 languages, 6,878 tools, 21,687 methods extracted at average precision 75%.