Research and Application of Question Answering System in the field of Air Traffic Control

Abstract

In this paper, we will build a question answering system for the field of air traffic control, which can assist controllers in monthly training better. The air traffic control question and answer system consists of a search based question answering system and a generation based question and answering system. The search based question answering system maintains an air traffic control problem set, uses word2vec to segment the questions and cosine theorem to calculate the similarity of the questions in order to match the best answer. To questions that can not be matched, we propose Attention-Based Bidirectional Long Short-Term Memory Networks to generate an answer.