Complex Question Answering over Large-scale Knowledge Bases

In this research study, we propose an effective framework for complex question answering over large-scale knowledge bases. We employ encoder-decoder neural networks, deep learning, reinforcement learning, meta-learning, few-shot learning algorithms, and other effective techniques to construct our question-answering framework to address the challenges existing in previous state-of-the-art question answering systems.  Empirical studies over large-scale CQA datasets not only indicate that our proposed approach is effective as it outperforms state-of-the-art methods significantly and also shed light on the role that specific components play in the question-answering task.