QQ客服 官方微博 反馈留言

	Learning Domain-specific Sentiment Lexicon with Supervised Sentiment-aware LDA
	Yang, Min; Zhu, Dingju; Mustafa, Rashed; Chow, Kam-Pui
	2014
会议名称	21st European Conference on Artificial Intelligence, ECAI 2014
会议地点	Czech Tech Univ, Prague, CZECH REPUBLIC
英文摘要	Analyzing and understanding people's sentiments towards different topics has become an interesting task due to the explosion of opinion-rich resources. In most sentiment analysis applications, sentiment lexicons play a crucial role, to be used as metadata of sentiment polarity. However, most previous works focus on discovering general-purpose sentiment lexicons. They cannot capture domain-specific sentiment words, or implicit and connotative sentiment words that are seemingly objective. In this paper, we propose a supervised sentiment-aware LDA model (ssLDA). The model uses a minimal set of domain-independent seed words and document labels to discover a domain-specific lexicon, learning a lexicon much richer and adaptive to the sentiment of specificdocument. Experiments on two publicly-available datasets (movie reviews and Obama-McCain debate dataset) show that our model is effective in constructing a comprehensive and high-quality domain-specific sentiment lexicon. Furthermore, the resulting lexicon significantly improves the performance of sentimentclassification tasks.
收录类别	EI
语种	英语
内容类型	会议论文
源URL	[http://ir.siat.ac.cn:8080/handle/172644/6068]
专题	深圳先进技术研究院_数字所
作者单位	2014
推荐引用方式 GB/T 7714	Yang, Min,Zhu, Dingju,Mustafa, Rashed,et al. Learning Domain-specific Sentiment Lexicon with Supervised Sentiment-aware LDA[C]. 见:21st European Conference on Artificial Intelligence, ECAI 2014. Czech Tech Univ, Prague, CZECH REPUBLIC.

个性服务

查看访问统计

相关权益政策

暂无数据

收藏/分享

除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。