A Bi-Directional LSTM-CNN Model with Attention for Aspect-Level Text Classification
2019-08-17 14:11
1601 查看
版权声明:本文为博主原创文章,遵循 CC 4.0 by-sa 版权协议,转载请附上原文出处链接和本声明。
本文链接:https://blog.csdn.net/qq_35709076/article/details/99683825
背景
属性级情感分析 (ABSA)是一项细粒度的情感分析任务,在自然语言处理里越来越受到关注,能够识别句子提及的属性以及对应的情感极性,包含两个子任务:属性识别和情感分析
例如:
“The food is not cheap but quite delicious”
实体food对应的属性:FOOD#PRICE和FOOD#TASTE
针对于人们越来越习惯在评论中使用习语,常会出现错误识别的情况,例如以下情况:
贡献“it is the service that leaves a bad taste in my mouth”
注意力机制能够识别出属性 “service”,然而没有注意力模型时,taste可能会被认为实体
1.提出融合注意力机制和属性信息的AARCNN模型。
2.注意力和属性是ABSA任务中必要的部分,因此介绍了两个模块:基于 CNN的注意力机制来抽取注意力和属性词权重embedding识别属性和分析。
3.实验结果表明该方法相比较几个 baseline方法,达到了最好的效果,并且进一步证明了 Bi-LSTM 提升了分类的准确率。
模型
1 Input Preprocess
首先需要将评论的句子进行格式化,假设句子是由n个单词序
相关文章推荐
- 【论文笔记】An End-to-End Model for QA over KBs with Cross-Attention Combining Global Knowledge
- The Application of Two-level Attention Models in CNN for Fine-grained Image Classification
- CIKM 2016 aNMM: Ranking Short Answer Texts with Attention-Based Neural Matching Model
- An unsupervised neural attention model for aspect extraction 读论文笔记
- An Analytical Model for a GPU Architecture with Memory-level and Thread-level Parallelism Awareness
- 行为识别阅读笔记(paper + parted code):Beyond Frame-level CNN Saliency-Aware 3-D CNN with LSTM for Video Acti
- Efficient Multi-Scale 3D CNN with fully connected CRF for Accurate Brain Lesion Segmentation 论文解读及实现
- 【SqueezeNet】SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <0.5MB model size
- aspect 分类及形容词的情感分类——An unsupervised aspect-sentiment model for online reviews
- Online Multi-Object Tracking Using CNN-based Single Object Tracker with Spatial-Temporal Attention M
- [NIPS 18] Stacked Semantics-Guided Attention Model for Fine-Grained Zero-Shot Learning
- 【论文阅读笔记】RETAIN: An Interpretable Predictive Model for Healthcare using Reverse Time Attention Mechan
- 情感分析论文阅读之《Aspect Level Sentiment Classification with Deep Memory Network》
- statistical-machine-learning-for-text-classification-with-scikitlearn-and-nltk
- Aspect-Oriented Programming with the e Verification Language: A Pragmatic Guide for Testbench Develo
- introduction with a Search Enghine for Text
- A generative vision model that trains with high data efficiency and breaks text-based CAPTCHAs(二)
- SIGIR 2016 Improving Language Estimation with the Paragraph Vector Model for Ad-hoc Retrieval
- Data Structure for HTML DOM with compare text inside
- data-intensive text processing with mapreduce-EM Algorithms for Text Processing