Abstract: Sentiment analysis has emerged as a widespread and effective technique for opinion mining of web data analysis. The development of the user-generated content has opened new prospects for research in the field of sentiment analysis. This paper analysis a model for sentiment analysis of twitter tweets using Unigram approach of Naïve Bayes. Firstly, tweets need to be downloaded using a free version tool called Node Xl. Once that is done Data pre-processing schemes are applied on the dataset. Secondly, the corpus needs to be trained. Thirdly, the behavior of Naive Bayes is studied in combination with different tweet topics to obtain the results for sentiment analysis.