Sentiment Analysis on Bangladesh Cricket with Support Vector Machine

Abstract

While social platform and news portal play a big role in Internet today, it also becomes the valuable medium for public opinions. We want to perform sentiment analysis in these public opinions. Many works have been done on sentiment analysis in different sectors for English language. But works in Bengali language are limited to only Bengali corpus and micro-blogging. So we have targeted a special sector which is Bangladesh Cricket where people express their opinions in their native Bengali languages on social medias in every moment. So we have prepared a dataset of three sentiment classes about Bangladesh Cricket from real people sentiments. We have processed our dataset by removing unnecessary words from the Bengali texts. Then we have used TF-IDF Vectorizer for vectorization and the classifier Support Vector Machine to classify our data.