Analyzing Social Emotions in Social Network Using Graph Based Co-Ranking Algorithm

2. Related Works

In this section analyze the various authors opinion involved in the sentiment examination process in social sites. Shulong Tan et al. (2014) proposes a Latent Dirichlet Allocation (LDA) based model. To distill foreground topics and filter out longstanding background topics, Foreground and Background LDA (FB-LDA) was used. To rank them with respect to their “popularity” within the variation period, Reason Candidate and Background LDA (RCB-LDA) was used. Alexandre Trilla et al. (2013) proposes the work focuses on categorization of a plain input text to inform a TTS system about the most appropriate sentiment (+ve, -ve, neutral) to automatically synthesize expressive speech at the sentence level. Alena Neviarouskaya et al. (2011) proposes the Latent Semantic Analysis (LSA) which assumes that words that are close in meaning will occur in similar pieces of text. Analyzing relationships between a set of documents and the terms contain in it. Parisa Lak et al. (2014) proposes the online product/service reviews serve as sources of product/service-related information star ratings provide a quick indication of tone of a review. In some cases, it is not available or detailed enough. Sentiment analysis automatically detect the polarity of text, i.e. more refined analysis.