Increasing competition and implementation of revenue management strategies in the airline industry have boosted accurate demand forecasting to maximize revenue. While most previous demand forecasting studies have focused on final booking demand of flights, this study develops a novel Transformer-based deep learning model to forecast flight advance bookings at daily granularity, named Flightformer. The proposed model considers the competition effect from other flights on the same route and determines the hyperparameters with Bayesian optimization. Subsequently, real flight advance booking data for nearly four years are used to evaluate the model performance. The results of the comparative experiment and ablation study show that Flightformer significantly outperforms the baseline and variant models. Moreover, the transferability of Flightformer on the other new route is examined.
The spread of rumors in social networks can do great harm to the society, so it is significant to limit the propagation of misinformation. One solution to block rumor is to broadcast the anti-rumor information. How to select users of social networks to spread information against rumor can be abstracted as the problem of influence blocking maximization (IBM). The problem can be described as choosing k nodes from a social graph to minimize the number of nodes that adopt rumor at the end of the spread process. The state-of-the-art IBM algorithm IBMM has a ( 1 - 1/e - ε)-approximation guarantee. However, we noticed that the algorithm could cause unnecessary cost of calculation. Therefore, we modify IBMM and propose the IBM-N algorithm. And the results of a series of experiments conducted on both synthetic and real world data sets demonstrate the performance efficiency of our algorithm.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.