Publication Type : Conference Paper
Publisher : International Conference on Energy, Communication, Data Analytics and Soft Computing.
Source : International Conference on Energy, Communication, Data Analytics and Soft Computing (ICECDS 2017) , Chennai, 2017.
Campus : Bengaluru
School : Department of Computer Science and Engineering, School of Engineering
Department : Computer Science
Year : 2017
Abstract : Nowadays, detecting the diseases that are widely spread in the society and predicting the future stages of the diseases has become very important in modern life. Twitter - a social media platform - a platform used by billions of users around worldwide to express their ideas about various topics, including health conditions. Twitter is used as source for public health information on global scale. To detect the disease and predict the future spread, social network system (SNS) is helpful. A model has been proposed for predicting the future trend of diseases (cancer) using twitter data. In this paper we built a single linear regression model by using tweets (total number of tweets and tweets related to influenza disease). Live streaming is done for finding the tweets from Forrest Gump website. However when data is incomplete the accuracy of prediction is poor. To improve the accuracy ridge regularization (eliminates the prediction error) is used. The data taken from the twitter is divided into training data and test data and through training data we predict death rates for the test data. When compared to prior studies our model is more accurate for the prediction of death rates.
Cite this Research Publication : A. P. Valli, .Uma, M., S.Pravallika, K. R., and Sasikala T, “Tracing out various diseases by analyzing twitter data applying data mining techniques”, in International Conference on Energy, Communication, Data Analytics and Soft Computing (ICECDS 2017) , Chennai, 2017.