A Scheme of Answer Selection In Community Question Answering Using Machine Learning Techniques

Abstract

Question Answering System is a very useful system in most of the deep learning related problems and it can be modeled as a question answering problem. Answer selection in Community Question Answering (CQA) is a challenging and important task to develop the automatic Question Answering System in a Natural Language Processing (NLP). The system requires semantic gap between questions answer pairs is the main problem in the QA system. It also needs a serious modeling of contextual factor. An attentive deep neural network architecture is proposed in this paper. The architecture is having three layers which are namely; Convolution Neural Network, Long Short Term Memory and Conditional Random Field. The SemEval-2015 CQA dataset is used to develop the experiment.