首页|PWEBSA: Twitter sentiment analysis by combining Plutchik wheel of emotion and word embedding
PWEBSA: Twitter sentiment analysis by combining Plutchik wheel of emotion and word embedding
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
Abstract Sentiment analysis is the task of predicting sentiments of words, sentences or entire document. This article addresses the issue of polarization of Twitter sentiments, which is one of the major areas of concern regarding sentiment analysis. The proposed approach consists of classifying the sentiments using emotions from Plutchik’s wheel of emotion, which provides eight basic emotions to make the tasks more approachable. To determine the polarity of texts, other features have been used as per Rule Based Emotion Classification (RBEM) algorithm. A decent amount of accuracy has been obtained using the proposed algorithm. Moreover, the classifier’s accuracy can be further increased if more words in the dictionary are added. The proposed approach has been compared with textblob API for sentiment analysis. This article discusses the technologies employed and contains details of tackling the issue of Sentiment Analysis through the proposed approach, thus justifying the increased accuracy over existing approaches. At the end, a glimpse of possible future work in this field has been included to show the researcher’s way ahead with the proposed approach.
Sentiment analysisNatural language processingWord embeddingPolarity
NETL
NSTL
