Retrieval-Based Factorization Machines for Human Click Behavior Prediction.

Human click behavior prediction is crucial for recommendation scenarios such as online commodity or advertisement recommendation, as it is helpful to improve the quality and user satisfaction of services. In recommender systems, the concept of click-through rate (CTR) is used to estimate the probability that a user will click on a recommended candidate. Many methods have been proposed to predict CTR and achieved good results.

[Effects of spacing on the yields and canopy structure of japonica rice at full heading stage].

With three panicle types of rice varieties and hybrids in Liaoning as entries, the effects of spacing of Japonica rice on light interception capacity, population light distribution, light conversion efficiency at full heading stage and yield were studied. The results showed that the leaf area indices at full heading stage, closely related to light interception, increased first and then decreased with the decrease of transplanting density. The extinction coefficient in one day increased first and then decreased, and the K value increased with the increase of planting density.