Enhanced Twitter Sentiment Analysis Using Hybrid Approach and by Accounting Local Contextual Semantic

Abstract

<jats:title>Abstract</jats:title> <jats:p>This paper addresses the problem of Twitter sentiment analysis through a hybrid approach in which SentiWordNet (SWN)-based feature vector acts as input to the classification model Support Vector Machine. Our main focus is to handle lexical modifier negation during SWN score calculation for the improvement of classification performance. Thus, we present naive and novel shift approach in which negation acts as both sentiment-bearing word and modifier, and then we shift the score of words from SWN based on their contextual semantic, inferred from neighbouring words. Additionally, we augment negation accounting procedure with a few heuristics for handling the cases in which negation presence does not necessarily mean negation. Experimental results show that the contextual-based SWN feature vector obtained through shift polarity approach alone led to an improved Twitter sentiment analysis system that outperforms the traditional reverse polarity approach by 2–6%. We validate the effectiveness of our hybrid approach considering negation on benchmark Twitter corpus from SemEval-2013 Task 2 competition.</jats:p>

Sentiment Analysis on COVID-19-Related Social Distancing in Canada Using Twitter Data

“…The lexicon-based approach, on the other hand, requires manual input of sentiment lexicons and performs well in any domain but fails to encompass complete informal lexicons. The hybrid approach will help to overcome the limitations of both approaches, thus enhancing performance, efficiency, and scalability [ 22 , 23 ]. Research has shown that using a hybrid approach not only accelerates accuracy and maintains stability but also provides better results than using one approach or one standard tool [ 23 , 24 ].…”

A Review on Negation Role in Twitter Sentiment Analysis

“…Nonetheless, critics such as Farooq et al (2017) and Gupta & Joshi (2019) have argued that the aforementioned methods might work well for simple text, but will fail in the case of compound and complex text (having more than one clause or conjunctions), owing to the presence of conjunction such as "but," which may limit the negation scope to only a single clause. As such, the scope of negation may possibly end before a static window termination or before the occurrence of the first punctuation mark.…”

Real-Time Twitter Corpus Labelling Using Automatic Clustering Approach

“…Thus, token "failure_NEG" would be extracted during feature engineering instead of "failure". http://journals.uob.edu.bh Furthermore, we exclude few negation tweets from the scope determination procedure which are having explicit negation words but literally, there is no sense of negation [15] as in tweet:…”