Σχετικά με το έγγραφο αυτό
- sl:arxiv_author :
- sl:arxiv_firstAuthor : Alexander Rietzler
- sl:arxiv_num : 1908.11860
- sl:arxiv_published : 2019-08-30T17:44:30Z
- sl:arxiv_summary : Aspect-Target Sentiment Classification (ATSC) is a subtask of Aspect-Based
Sentiment Analysis (ABSA), which has many applications e.g. in e-commerce,
where data and insights from reviews can be leveraged to create value for
businesses and customers. Recently, deep transfer-learning methods have been
applied successfully to a myriad of Natural Language Processing (NLP) tasks,
including ATSC. Building on top of the prominent BERT language model, we
approach ATSC using a two-step procedure: self-supervised domain-specific BERT
language model finetuning, followed by supervised task-specific finetuning. Our
findings on how to best exploit domain-specific language model finetuning
enable us to produce new state-of-the-art performance on the SemEval 2014 Task
4 restaurants dataset. In addition, to explore the real-world robustness of our
models, we perform cross-domain evaluation. We show that a cross-domain adapted
BERT language model performs significantly better than strong baseline models
like vanilla BERT-base and XLNet-base. Finally, we conduct a case study to
interpret model prediction errors.@en
- sl:arxiv_title : Adapt or Get Left Behind: Domain Adaptation through BERT Language Model Finetuning for Aspect-Target Sentiment Classification@en
- sl:arxiv_updated : 2019-11-19T10:17:52Z
- sl:bookmarkOf : https://arxiv.org/abs/1908.11860
- sl:creationDate : 2021-10-21
- sl:creationTime : 2021-10-21T12:56:49Z
Προτιμήσεις...