Comparative study for machine learning classifier recommendation to predict political affiliation based on online reviews

In the current era of social media, different platforms such as Twitter and Facebook have frequently been used by leaders followers political parties to participate in events, campaigns, elections. The acquisition, analysis, presentation content received considerable attention from opinion-mining researchers. For this purpose, supervised unsupervised techniques used. However, they produced less efficient results, which need be improved incorporating additional classifiers with extended data sets. authors investigate machine learning for classifying affiliations users. a set reviews is acquired annotated polarity classes. After pre-processing, like K-nearest neighbor, naïve Bayes, support vector machine, extreme gradient boosting, others, are applied. Experimental results illustrate that boosting shown promising predicting affiliations.