Mining public opinion: a sentiment based forecasting for democratic elections of Pakistan

Abstract

Twitter has emerged as outstanding and most prominent social media in today's technological age. The data proliferates in quick and words with its activities trigger get fast responses from the users. This platform is perfect for promoting political perspectives, particularly when election campaigns are on its peak. Political trends on Twitter media has been contemplated in the course of recent years. In the past research, both supervised and unsupervised methodologies have been used to analyze the Twitter trends. Most of the Tweet classification approaches utilized built in Dictionaries, Naı ¨ve Bayes, K-Nearest Neighbors (KNN), decision tree and Support Vector Machines (SVM) classifiers. However, in case of democratic election these trends can be mined to predict the winning party. However, all such approaches produce poor results due to language issues, low accuracy, limited access to internet and lower literacy rate in less developed countries such as Pakistan. This research study, find the best possible way for collection of tweets related to different political parties and build a prediction model that may analyze sentiments and opinions expressed by peoples in their Tweets. In this research work, a prediction based model along with novel similarity measure has been proposed to predict the election results of political parties in Pakistan. The proposed work is composed of data collection, preprocessing, aspect extraction, aspect refinement and final prediction using Bayesian theorem. Form the experimental results, it is concluded that proposed approach perform better than existing techniques by obtaining almost 98% accuracy and efficiently cover the limitations of existing studies.

Aspect-Based Sentiment Analysis for Social Multimedia: A Hybrid Computational Framework

“…In addition, the Stanford Dependency Parser (SDP) approach is utilized to extract implicit aspects, which consider the relationship between the opinion and aspects. Rule mining approaches are very much used in past literature for sentiment analysis [13,19,35]. The following sections go into detail about each of these components.…”

“…The polarity intensity of emotion words is not considered by existing approaches, which divide sentiment words into negative and positive categories. In some instances, current methods are domain-specific [17]; when applied to a dataset from a different domain, these approaches lose their accuracy [18,19]. The following is a list of the research's key findings:…”

Anatomy of Sentiment Analysis of Tweets Using Machine Learning Approach

“…Twitter has more than 650 million registered users and it is commonly ranked as one of the most popular online social networking web site, although practically, it is the third most popular after the Instagram and Facebook [2]. Through online groups, one can easily join media where consumers notify and bias something through the forums [1,3]. Due to the vast usage of social media forums, it has been observed that a huge volume of sentiment-rich data within the realm of tweets, status upgrades, blog publish, remarks, and reviews are being generated at every movement.…”

“…Due to the vast usage of social media forums, it has been observed that a huge volume of sentiment-rich data within the realm of tweets, status upgrades, blog publish, remarks, and reviews are being generated at every movement. Moreover, social media gives a chance to various stakeholders such as businesses by giving a floor to connect with their customers for advertising and dealings [3]. Common people, on the whole, may also utilise the online user-created content to the best length for decision making.…”