| Autor: |
Gribkov, E. I., Yekhlakov, Yu. P. |
| Zdroj: |
Scientific & Technical Information Processing; Dec2021, Vol. 48 Issue 5, p452-460, 9p |
| Abstrakt: |
Extraction and analysis of user opinions towards products and services is important task in research and applications of natural language processing methods. We frame this task as a structured prediction task where each data instance is represented by a group of interdependent labels. To solve this task, we describe a transition-based model that decomposes it to the prediction of a transition sequence that incrementally builds the final structure. The proposed model uses deep neural networks as feature extractor for a classifier that predicts the next transition based on previous transitions and the parts of predicted structure. To evaluate the quality of the proposed model, we conducted a series of experiments on user reviews texts from two sources: English reviews from Amazon and Russian reviews from AliExpress. The experiments show that our model performs equally or better than an alternative and suffers of an less accuracy drop from a distribution shift. [ABSTRACT FROM AUTHOR] |
| Databáze: |
Complementary Index |
| Externí odkaz: |
|
Full text from SpringerLink