Sentiment analysis based on twitter

Sentiment analysis has become more crucial after the rise of social media, especially for the Twitter since it provides structured and publicly available data. TwitterSentiDetector is a domain-dependent and unsupervised Twitter sentiment analyzer that focuses on the differences occurred by the informal language used in Twitter such as spelling mistakes, letter repetitions, usage of hashtags, emoticons, emojis, and laughs. TwitterSentiDetector uses natural language processing techniques alongside the proposed linguistic methods to classify sentiments of tweets into positive, negative, and neutral through the polarity scores obtained from widely used sentiment lexicons. According to tests on the widely-used Twitter datasets that contain manually detected sentiment labels alongside tweets, TwitterSentiDetector's sentiment detection ratio is calculated as up to 69%. When the target sentiment classes are decreased to positive and negative, the detection ratio is increased up to 87%. The results are calculated very similarly when the same dataset is evaluated by the proposed tweet-level context aware sentiment analysis module which confirms the validity of each approach. A Twitter sentiment analyzer services should be aware of spam since it is still widespread in Twitter and one of its aim is to hijack the validity of the services based on Twitter. According to the experimental results, the integrated spam detection framework's accuracy is calculated as 0.943. Similarly, when the developed graph based sentence level spell checking framework which is integrated into TwitterSentiDetector is evaluated using the most commonly misspelled words in English which are based on three lists, the accuracy is calculated as 0.84.