The amalgamation of technology, social media, and innovation has a significant influence on human life. Most of us spend their time on social networking sites for their decision making. An opinion leader has enormous significance for human's decision-making process. In this paper, we addressed an inventive and novel approach to discovering the opinion leader based tweets posted by the user on the respond of the query posed by another user in the social network. We used the sentiment analysis based model to classify the tweets/ reviews on a particular topic in the data set. We calculated the positive, negative, and neutral score and measured the polarity of each statement using the NLTK python package. Next, we calculated the total polarity cost and ranked the user based on the total polarity cost. The user has a higher polarity cost considered as an opinion leader. In this paper, we have implemented the proposed model on Amazon Fine Foods reviews data set having around 568,454 reviews posted by 256,059 users. For validating the proposed model, we compared the results with the other standard classifier and found that the proposed model is better than another classifier in terms of accuracy, precision, recall, and F1-score.