Abstract:

Abstract In the starting of year 2020, WHO identified COVID-19 as a new pandemic and issued a statement to that effect. This fatal virus was able to disperse and propagate over several nations all over the globe. During the course of the epidemic, social media platforms like Twitter generated significant and substantial volumes of data that helped improve the quality of decisions pertaining to health care. As a result, we suggest that the opinion expressed by users might be analysed via the use of efficient Supervised Machine Learning (SML) algorithms to forecast the occurrence of illness and offer early warnings. In this paper we proposed a text mining classifier for generate the summarized text using machine learning techniques.  After collecting the tweets, we got them ready for pre-processing and generate the class label for all instances such as correct, incorrect and neutral etc. In the second phase, numerous features are extracted from text by using a number of frequently used approaches, such as TF-IDF, co-relational, NLP and relational dependency features are extracted to generate the feature vector. As classification module we used one binary classification algorithm and five machine learning algorithms for evaluation of proposed model. NLP-SVM and TFIDF-SVM produces higher accuracy 95.10% and 93.50% classification accuracy respectively. This demonstrates the proposed model is effective for classification of large text for COVID-19 on tweet data.

Keywords: