Recurrent convolution neural networks for classification of protein-protein interaction articles from biomedical literature

Text classification (TC) is a task that assigns a text to one or more classes and predefined categories. Constructing text classifiers with high accuracy is a vital task in biomedical field, given the wealth of information hidden in unlabelled documents. Because of large feature spaces, traditionally discriminative approaches, such as logistic regression and support vector machines with n-gram and semantic features have been utilizedfor biomedical text classification. In this study, we propose Recurrent Convolution Neural Networks (RCNN) based automated technique for classifying protein-protein interaction (PPI) articles. In RCNN model we utilized a recurrent structure to detain the contextual information from word embedding features. Max pooling layer was configured to extract important semantic keywords from the text. We evaluated our approach on two benchmark PPI datasets BioCreative II and BioCreative III. An experimental results show that RCNN based protein-protein interaction classification approach performs better than other state of the art approaches.