Sanjeev Ahuja  used the SentiWordNet for obtaining the overall polarity of the movie review tweets which will get the tweet data information form WordNet. It includes sentiment of words and comprises into sentiment scores with positive, negative and also objective scores. Calculation of all these scores gives the relative strength into it. The State of Art approach has been included for increasing the performance and accuracy of the tweet data.
Rachana Bandana,  proposed the sentiment analysis with document level for the movie review classification applied by heterogeneous features, for supporting the supervised machine learning algorithms as Naive Bayes and Linear Support Vector Machine. With these the author learns and classified the text reviews into positive and also a negative list.
Purtata Bhoir,  used the subjectivity analysis with two methods as SentiWordNet and Naive Bayes. Among these Naive Bayes classifier is a probabilistic model theorem to calculate the probability with a class of individual sentences or text. Naive Bayes analyzed for extracting the feature and opinion from the public tweets and gives better solution compared to SentiWordNet.
M. Ali Fauzi,  shows the ensemble technique with Naive bayes for sentiment analysis of movies on Indonesian Twitter data. The ensemble features includes twitter specific features, textual features, part of speech features, and lexicon-based features, and Bag of Words. They came up with lexicon-based features which hold the percentage of positive and negative tweet words based on their part of speech.
V.K. Singh, R. Piryani, A. Uddin,  introduces the sentiment classifiers also called as IR formulations are used not only for reviewing the positive and negative tweet command, also used for highlighting the tweet products with a very short period of time used in different domains.
Vasilija Uzunova,  proposed sentiment analysis of film reviews in Macedonian using Naive Bayes. With the help of sentiment polarity they classified the film comment text and checked whether it is positive or negative based on opinion of the public reviews.
H. Ghorbel and D. Jacot,  introduced the sentiment analysis for French movie reviews with the help of shallow linguistic features. Three different categories are performed for this movie reviews. One is “lexical” approach determines from the word unigrams and it find some similar tweets in positive and negative reviews. Another one as “morpho-syntactic” approach which reduces the tweet features into part of speech categories for improving the performance of data. Then the “semantic” approach uses the lexical feature of sentiwordnet to increase the polarity of tweet word and calculating the overall polarity score of the movie review for each part of speech tag.
Liang-Chu Chen et al,  proposed the sentiment analysis of social network for Military life board on largest online communities in Taiwan. The text mining occupied some systematic method for this online community along with the help of self-organized military sentiment vocabulary. A web crawler method used for this military life PPT board to extract the content of military posts and message blogs and also analyze this online community which calculates it from the different parameters.
Ishan Arora,  introduced the Naive Bayes classifier for calculating the conditional independence of tweet class as positive or in a negative form and the tweet words are conditionally independent with each other. With the help of this the accuracy of the tweet data classification will not get any issue and it gives a fastest performance for the classification problem.
Deepa Anand and Deepan Naorem,  works on filtering the sentiment analysis statements from the public reviews and grouping it with a corresponding aspects of categories. The review of sentence filtering is used to detect the accuracy of sentiment tweets and it is different from the subjectivity classification. The tweet sentence problem denoting a target of public other than the items being reviewed for the movie.
Barkha Bansal et al,  works on demonstrating the CBOW model compared with skip- gram model for customer reviews in mobile phone application. The semantic features using the cosine distance measure, for each input of mobile phone.