An LSTM Based Deep Learning Method for Airline Ticket Price Prediction

Airline ticket prices are changing all the time dynamically. Booking tickets with the lowest costs is a challenging task. In this paper, we propose a multi-layer convolution long short-term memory network model (MLC-LSTM) and associated time series based data processing method for airline ticket price prediction. In our model, the parallel fully connected LSTM blocks extract the independent historical data from different flights and the following convolution layers merge and decode the inter and intra-information. From the results of experiment, the proposed model outperforms many existing models such as the basic LSTM, and the efficiency of the data processing method is also proved.