文章基本信息

标题：A Statistical Method of Knowledge Extraction on Online Stock Forum Using Subspace Clustering with Outlier Detection
本地全文：下载
作者：N.Pooranam ; G.Shyamala
期刊名称：International Journal of Innovative Research in Science, Engineering and Technology
印刷版ISSN：2347-6710
电子版ISSN：2319-8753
出版年度：2016
卷号：5
期号：5
页码：7249
DOI：10.15680/IJIRSET.2016.0505087
出版社：S&S Publications
摘要：Financial stock Data Analysis and future prediction in terms of Sentiments is great challenge in the bigdata research. Among the unlabeled opinion, opinion classification in terms of unsupervised learning algorithm willlead to classification error as data is sparse and high dimensional. To overcome this problem, the sentiment analysis toextract the opinion of each word in the stock data has been proposed. Moreover the data size is large, hence the singularvalue decomposition to resolve the inconsistent constraints correlating to the large dimensions, and dimensionallyreduced feature set is been used. The dimensionally reduced feature set is classified into clusters through employmentof Principle component analysis which is used to calculate the strength of actionable clusters with utilization of thedomain knowledge and validate the optimal centroids dynamically. Cluster data which further inconsistent with theoutlier probability can further reduced through subspace clustering. Experimental results prove that the proposedframework outperforms the state of art approaches in terms of precision, recall and F measure.
关键词：Opinion mining; sentiment analysis; Singular Value Decomposition; Principle Component Analysis;Subspace Clustering