In this paper, we propose a new neural network architecture for item recommendation with structural information. Our model, structural neural recommender (SNR) is based on neural networks and operates on a hierarchy paradigm, aiming to explore the effectiveness of incorporating different structural information for recommendation. Many recent state-of-the-art neural network based recommendation models exploit the nonlinear transformations for modeling the complex user-item interaction patterns and user historical behaviors, ignoring the item-item structural relationship. This structural information, however, is uncomplicated to derive and useful for inferring item characteristics. To utilize this information, SNR simultaneously learns representation from user-item interactions and item-item relationships. Empirical studies on eight real-world datasets demonstrate the effectiveness of incorporating such structural information, by outperforming classic and recent baselines. We also conduct detail ablation studies and hyper-parameter analysis to provide further understanding towards the behaviors of our model. Following the model development, we conduct a field experiment to demonstrate that the effectiveness of algorithmic recommender systems can further increase by using different types of message framing when communicating recommendations to consumers. Our results suggest that recommendations framed with a relevance appeal (e.g. “Top 5 brands for you”) are more effective in general, yet recommendations that are framed with a popularity appeal (“Top 5 most popular brands”) are more effective for customers who were acquired via social media (versus non-social media) advertising or for those who have stronger (versus weaker) social orientation.

The warm recycling technology has been increasingly used in many countries due to the environmental and financial benefits. In this study, the rheological and fatigue performance evolutions of warm-mix recycled asphalt materials during the secondary service period were evaluated in two scales, mixture and fine aggregate matrix (FAM). A laboratory simulation method was proposed to produce warm-mix recycled asphalt binders with various long-term aging levels for the mixture and FAM tests. The dynamic shear rheometer temperature and frequency sweep test and time sweep test were conducted to characterize the rheological and fatigue behavior of FAMs, respectively. The rheological and fatigue properties of asphalt mixtures were measured by the dynamic modulus test and semi-circular bending test, respectively. Effects of aging levels and recycling plans on different pavement performance were investigated. Performance correlations between the mixture and FAM were finally investigated by the statistical method. It is found that the secondary long-term aging causes the continuous increase in the stiffness and decrease in the viscoelasticity level in each material scale, indicating the improvement of the rutting resistance and the reduction of the fatigue resistance. The warm mix asphalt technology plays a positive role in the fatigue performance with a loss of the rutting resistance. Using the styrene butadiene rubber latex can improve different pavement performance within the whole time-temperature domain. Good performance correlations between the mixture and FAM are developed, indicating that the FAM may be the critical material scale for evaluating the rheological and fatigue performance of warm-mix recycled asphalt pavements.