The sentiment analysis is a process which helps to categorize the people’s opinion expressed in customers’ reviews which determines the customers’ attitude towards any service, or product is positive, neutral or negative. It is mostly used in social media monitoring to allow practitioners and researchers to gain wider public (Indian consumers in Indian market scope) opinions overview behind social networks, product review, etc. This research is interested in identifying more in expressed public sentiments (positive, negative or neutral) towards features of product rather to identify sentiments in product review texts. There are several supervised machine learning methods that have been conducted in the past to calculate product features sentiment scores and sentence-level or document-level sentiment analysis respectively. In this research work, the sentiment or polarity scores of features of product are calculated with the help of rulebased Python Text Blob library. Product feature based sentiment analysis is conducted by supervised machine learning methods. The feature extraction method, TF-IDF method is applied. Finally the result of this research shows (i) the features of product became good indicator during determination of polarity or sentiment classifications of review texts, (ii) Random Forest supervised machine learning performed well with 92% accuracy in polarity classification with the hyper parameter Information Gain.