Aspect Term Extraction for Sentiment Analysis: New Datasets, New Evaluation Measures and an Improved Unsupervised Method
John Pavlopoulos, Ion Androutsopoulos · 2014
Given a set of texts discussing a particular entity (e.g., customer reviews of a smart-phone), aspect based sentiment analysis (ABSA) identifies prominent aspects of the entity (e.g., battery, screen) and an aver-age sentiment score per aspect. We fo-cus on aspect term extraction (ATE), one of the core processing stages of ABSA that extracts terms naming aspects. We make publicly available three new ATE datasets, arguing that they are better than previously available ones. We also introduce new evaluation measures for ATE, again argu-ing that they are better than previously used ones. Finally, we show how a pop-ular unsupervised ATE method can be im-proved by using continuous space vector representations of words and phrases. 1