Looking for trouble

Abstract

This paper presents a method for mining potential troubles or obstacles related to the use of a given object. Some example instances of this relation are (medicine, side effect) and (amusement park, height restriction). Our acquisition method consists of three steps. First, we use an unsupervised method to collect training samples from Web documents. Second, a set of expressions generally referring to troubles is acquired by a supervised learning method. Finally, the acquired troubles are associated with objects so that each of the resulting pairs consists of an object and a trouble or obstacle in using that object. To show the effectiveness of our method we conducted experiments using a large collection of Japanese Web documents for acquisition. Experimental results show an 85.5% precision for the top 10,000 acquired troubles, and a 74% precision for the top 10% of over 60,000 acquired object-trouble pairs. © 2008 Licensed under the Creative Commons.