Personalizing large-scale text classification by modeling individual differences

Jungho Lee, Byung-Ju Choi, Yeachan Kim, Kang-Min Kim, Woo-Jong Ryu, SangKeun Lee

Abstract: Large-scale text classification is used to organize and subsequently, analyze textual information into a variety of topics effectively. However, most of existing large-scale text classification models tend to draw similar classification results without accounting for the differences in individual perceptions, as may be discernable through the text semantics based on distinct human characteristics. In this paper, we propose a personalized large-scale text classification model, which factors in these individual differences when classifying data.