Naives Bayes Algorithm for Twitter Sentiment Analysis

Abstract

On 2 March 2020, the Indonesian government, through President Joko ‘Jokowi’ Widodo, announced the first two cases of COVID-19 in Indonesia. This is the first case of COVID-19 officially confirmed in that country. Several cases have continued to increase since then. President Jokowi began issuing policies on the spread of this virus. This is different from other countries, such as Malaysia and Singapore, which responded from the previous month when the Indonesian government still stated that coronavirus does not exist in Indonesia. Our case study is to find a public opinion through social network analysis of Indonesian public policy during the beginning of the Indonesian COVID-19 pandemic in March 2020. This research implements text mining and document-based sentiments on Twitter data that is reprocessed through machine learning techniques using the Naïve Bayes method. We found negative opinions in the period more dominant by 46%, while that was 35% positive sentiment and 20% neutral. This research shows that anticipation, sadness, and anger are very dominant in the emotional analysis.

Sentiment analysis of face-to-face learning based on social media

“…Before applying the algorithm, the preparation stage is necessary. Case folding and URL removal are performed at this stage to uniformize the letters used into lowercase letters and delete the URL connection, as illustrated in Table 2 (Samsir et al, 2021). The preprocessed data is then sorted for the most word frequencies to emerge.…”

“…Their widespread adoption and importance in modern life are among the most intriguing developments in recent years, attracting the attention of scholars, journalists, businesses, and governments. The overall tendency in sentiment analysis research in social networks is to utilize techniques acquired from classical sentiment analysis explored since the early 2000s (Samsir et al, 2021).…”

“…The cleaning procedure is carried out at this step to decrease noise and remove stopwords to remove useless words. Tokenization is a method that uses spaces and punctuation to identify words and divide sentences down into terms (Samsir et al, 2021). The next stage of preprocessing is stemming, which converts affixes into fundamental words.…”

Implementation Naïve Bayes Classification for Sentiment Analysis on Internet Movie Database

“…The naive Bayes classification method is based on Bayesian theory and uses statistical methods to predict the likelihood of membership in each class. Based on the Nave Bayes assumption, which states that the attribute value of one class will not impact the attribute value of another class (class-conditional independence) [2] [15]. The total probability distribution is thus the product of the probability distributions of each data tuple, as illustrated in the equation below:…”

Sentiment analysis on vaccine COVID-19 using word count and Gaussian Naïve Bayes

“…Sentiment evaluation has regularly been finished with the aid of using associated research. For example, for sentiment evaluation on COVID-19, picture sentiment airline reviews, political sentiment, FB comment, resort reviews, PC reviews, patron satisfaction/reviews, training e-sport, film reviews, polygamy or even sentiment evaluation may be used for product and carrier evaluation [11]- [20].…”