Employ Twitter Data to Perform Sentiment Analysis in the Malay Language

Abstract

The main intention of this research is to discover the significance of sentiment analysis of Twitter data whether it is positive, neutral or negative. The sentiment analysis is dependent mining about textual content which being extracted and identified as contextual and subjective knowledge in such perceptible origin of recent rapid expanding computer science research. We started with a systematic literature review, where we had adopted both qualitative coding and text mining by scrutinizing 3282 of input of textual data retrieved from Twitter Streaming API. We perceived the problem as the decision trees kind of sentiment analysis in learning and information gaining. Therefore, we showed how basic decision trees are built to calculate the sentiment values of Twitter data. Sentiment analysis has transformed from interpreting online textual output analysis into perceiving contextual social media texts for example from Twitter. Hence, two decision trees were built to observe the performance and information gaining of decision trees. Thus, the precision of both decision trees led to the precision percentage that will be respectively stated, and the best decision tree can be obtained.