作者: Jay Kuan-Chieh Chung , Chi-En Wu , Richard Tzong-Han Tsai
DOI: 10.1007/978-3-319-12024-9_7
关键词: Natural language processing 、 Task (project management) 、 Polarity (physics) 、 Computer science 、 Knowledge base 、 Machine learning 、 Artificial intelligence 、 Set (abstract data type) 、 Sentiment analysis
摘要: In this paper, we present our system that participated in the Polarity Detection task, elementary task ESWC-14 Challenge on Concept-Level Sentiment Analysis. addition to traditional Bag-of-Words features, also employ state-of-the-art Sentic API extract concepts from documents generate Bag-of-Sentiment-Concepts features. Our previous work SentiConceptNet serves as reference concept-based sentiment knowledge base for concept-level analysis. Experimental results development set show adding can improve accuracy by 1.3 %, indicating benefit of demo website is located at http://140.115.51.136:5000.