Dictionary Based Gender Identification and Gender Based Sentiment Analysis with Polarized Word2Vec

ABSTRACT

Each gender is having its special behaviour which can be reflected in every field of social media. During the pandemic of COVID-19, people used twitter to discuss the issues caused by COVID-19 disease. As Twitter does not disclose the gender of the user, in this study we have discussed different kinds of approaches used to identify the gender. From the literature review, it is found that the dictionary-based approaches are the best suitable approach when we are working with the sentiment analysis of unlabelled data. This study is about the analysis of ten kinds of emotions of males and females by which we can observe how they reacted in this pandemic. The research proposes a dictionary-based approach to identify the gender and then analyzed sentiments using the cluster-based approach is applied onto word vectors after multiplying them with sentence’s polarity. The proposed approach is compared with the existing approaches with different data set and found that our proposed approach depicts good accuracy of sentiment analysis of unlabelled gendered data. © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.