استخراج الرأي المستهدف واستخراج الأجل رأي هما مهمتان أساسيتان في تحليل المعرفات القائم على الجانب (ABASA). تركز العديد من الأعمال الأخيرة على ABSA على استخراج كلمات الرأي الموجهة نحو الهدف (TOWE) (Towe)، والتي تهدف إلى استخراج كلمات الرأي المقابلة للحصول على هدف رأي معين. يمكن تطبيق Towe أكثر استخداما على استخراج زوج الرأي في الجانب (AOPE) الذي يهدف إلى استخراج الجوانب (أي أهداف الرأي) وشروط الرأي في أزواج. في هذه الورقة، نقترح تسلسل تسلسل محدد بالهدف مع اهتمام الذات متعدد الرأس (TSMSA) ل Towe، حيث يمكن دمج أي نموذج لغة مدرب مسبقا مع اهتمام ذاتي متعدد الرأس بشكل مريح. كدراسة حالة، نقوم أيضا بتطوير هيكل متعدد المهام يدعى MT-TSMSA من أجل الجمع بين TSMSA لدينا مع وحدة استخراج الجانب والرأي. تشير النتائج التجريبية إلى أن TSMSA تتفوق على الأساليب القياسية على Towe بشكل ملحوظ؛ وفي الوقت نفسه، فإن أداء MT-TSMSA متشابه أو حتى أفضل من نماذج خط الأساس الحديثة.
Opinion target extraction and opinion term extraction are two fundamental tasks in Aspect Based Sentiment Analysis (ABSA). Many recent works on ABSA focus on Target-oriented Opinion Words (or Terms) Extraction (TOWE), which aims at extracting the corresponding opinion words for a given opinion target. TOWE can be further applied to Aspect-Opinion Pair Extraction (AOPE) which aims at extracting aspects (i.e., opinion targets) and opinion terms in pairs. In this paper, we propose Target-Specified sequence labeling with Multi-head Self-Attention (TSMSA) for TOWE, in which any pre-trained language model with multi-head self-attention can be integrated conveniently. As a case study, we also develop a Multi-Task structure named MT-TSMSA for AOPE by combining our TSMSA with an aspect and opinion term extraction module. Experimental results indicate that TSMSA outperforms the benchmark methods on TOWE significantly; meanwhile, the performance of MT-TSMSA is similar or even better than state-of-the-art AOPE baseline models.
References used
https://aclanthology.org/
Multi-head self-attention recently attracts enormous interest owing to its specialized functions, significant parallelizable computation, and flexible extensibility. However, very recent empirical studies show that some self-attention heads make litt
Sequence labeling aims to predict a fine-grained sequence of labels for the text. However, such formulation hinders the effectiveness of supervised methods due to the lack of token-level annotated data. This is exacerbated when we meet a diverse rang
We investigate how sentence-level transformers can be modified into effective sequence labelers at the token level without any direct supervision. Existing approaches to zero-shot sequence labeling do not perform well when applied on transformer-base
Following the increasing performance of neural machine translation systems, the paradigm of using automatically translated data for cross-lingual adaptation is now studied in several applicative domains. The capacity to accurately project annotations
Current benchmark tasks for natural language processing contain text that is qualitatively different from the text used in informal day to day digital communication. This discrepancy has led to severe performance degradation of state-of-the-art NLP m