Implicit and Explicit Aspect Extraction in Financial Microblogs
Refereed Conference Meeting Proceeding
This paper focuses on aspect extraction which is a sub-task of Aspect-based Sentiment Analysis. The goal is to report an extraction method of financial aspects in microblog messages. Our approach is based on a stock-investment taxonomy for the extraction of explicit and implicit aspects. We compare supervised and unsupervised methods to assign stock-investment aspect categories to microblog messages. Our method targets implicit and explicit aspects. Results on 7 aspect classes reach 0.71 accuracy. Improvements are required in the case of 32 class classification, especially concerning implicit aspects.
Annual Meeting of the Association for Computational Linguistics)
Proceedings of the First Workshop on Economics and Natural Language Processing
Digital Object Identifer (DOI):
National University of Ireland, Galway (NUIG)
Open access repository: