LT3: Applying Hybrid Terminology Extraction to Aspect-Based Sentiment Analysis
Orphée De Clercq, Marjan Van de Kauter, Els Lefever, Véronique Hoste · 2015
The LT3 system perceives ABSA as a task consisting of three main subtasks, which have to be tackled incrementally, namely aspect term extraction, aggregation and polarity classification.For the first two steps, we see that employing a hybrid terminology extraction system leads to promising results, especially when it comes to recall.For the polarity classification, we show that it is possible to gain satisfying accuracies, even on out-ofdomain data, with a basic model employing only lexical information.