<![CDATA[With the rapid growth of social media in recent years, opinion mining, also called Sentiment Analysis, has gained much attention. Opinion mining refers to the use of Natural Language Processing (NLP), text analysis and biometrics to analyse, extract and identify the customer views of a product. The polarity classification is a main task of opinion mining for getting useful decision from online customer reviews and survey responses. Existing approaches in opinion mining performs the classification based on the parametric form. In this paper, we introduce the Naive Bayes classifier with Kernel Density Estimation (KDE) as a non parametric way of opinion mining that computes the probability density function based on the kernel estimator. The features are extracted from the opinions in the document level and are used to train the classifier in a supervised manner. KDE also performs data smoothing problem that inferences about the population that are based on finite data sample. We underwent some pre-processing techniques like stemming and stopword removal. Experimental results show that the proposed algorithm performed well, in terms of accuracy, precision, recall and F-measure.]]>
Copyrights © 2022
