Applying machine learning techniques to large volumes of text at the big data level requires a large amount of training data and several studies are being introduced to create automatic (or semi-automatic) machines. In this paper, we devise and develo...
Applying machine learning techniques to large volumes of text at the big data level requires a large amount of training data and several studies are being introduced to create automatic (or semi-automatic) machines. In this paper, we devise and develop a technique for automatically labeling large-volume online news articles produced by hundreds of media in Korea. From the experimental results, we could find that the accuracy of the predicted results is more than 92%.