Published May 24, 2018
| Version v1
Conference paper
Open
Implicit and Explicit Aspect Extraction in Financial Microblogs
Creators
- 1. nsight Centre for Data Analytics NUI Galway
Description
This paper focuses on aspect extraction which is a sub-task of Aspect-based Sen- timent Analysis. The goal is to report an extraction method of financial aspects in microblog messages. Our approach uses a stock-investment taxonomy for the iden- tification of explicit and implicit aspects. We compare supervised and unsupervised methods to assign predefined categories at message level. Results on 7 aspect classes show 0.71 accuracy, while the 32 class classification gives 0.82 accuracy for mes- sages containing explicit aspects and 0.35 for implicit aspects.
Files
econlp2018-workshop-implicit.pdf
Files
(127.1 kB)
Name | Size | Download all |
---|---|---|
md5:84483c44f6691c1a695e5af26f9918dd
|
127.1 kB | Preview Download |