Title :
Restricted Domain Opinion Mining in Compound Sentences
Author :
Mishra, Nitesh ; Jha, Chandan Kumar
Author_Institution :
Comput. Sci., Banasthali Vidyapith, Vanasthali, India
Abstract :
The growing availability of opinion rich resources like online review sites and blogs arises as people can easily seek out and understand the opinions of others. Users express their views and opinions that could be simple or compound sentence which is related to the products and services. These opinions are subjective information which represents user´s sentiments, feelings or appraisal related to the same. With respect to compound sentences there are number of problems in opinion mining such as identification of object features, their individual sentiments and sentiments of overall object as a whole. This paper is an extension to our work on semantic question answering in which we identify the score of the asked feature by deriving it from the question through opinion mining in compound sentences. Our method shows significant improvement on movie-review dataset.
Keywords :
Web sites; data mining; natural language processing; question answering (information retrieval); blog; compound sentence; movie-review dataset; object feature identification; online review site; opinion rich resource; restricted domain opinion mining; semantic question answering; Compounds; Data mining; Feature extraction; Knowledge discovery; Motion pictures; Semantics; Sentiment analysis; Machine learning; Opinion Mining; Polarity; SentiWordNet; Sentiments; semantic;
Conference_Titel :
Communication Systems and Network Technologies (CSNT), 2013 International Conference on
Conference_Location :
Gwalior
Print_ISBN :
978-1-4673-5603-9
DOI :
10.1109/CSNT.2013.132