Anti-Proinsulin monoclonal antibody, clone 3I6

The rapid development of location-based social networks (LBSNs) produces the increasing number of check-in records and corresponding heterogeneous information which bring big challenges of points-of-interest (POIs) recommendation in our daily lives. The emergence of various recommender techniques bridges the gap between the numerous heterogeneous check-ins and the personalized POI recommendation. However, due to the differences between LBSNs and conventional recommendation tasks, besides the user feedback, the spatio-temporal information is also significant to precisely capture the user preferences. In this paper, we propose a multi-task learning model based POI recommender system which exploits a structure of generative adversarial networks (GAN) simultaneously considering temporal check-ins and geographical locations. The GAN-based model is capable of relieving the sparsity of check-in data in POI recommender systems. The temporal check-ins not only present the preference but also show the lifestyle of an individual while the geographical locations describe the active region of users which further filters POIs far from the feasible region. The multi-task learning strategy is capable of combining the information of temporal check-ins and geographical locations to improve the performance of personalized POI recommendation. We conduct the experiments on two real-world LBSNs datasets and the experimental results show the effectiveness of our proposed approach.

The instantaneous kinematic velocity obtained by using a stand-alone global navigation satellite system (GNSS) receiver has attracted increasing attention, and has significant for numerous applications. Using the time difference carrier phase (TDCP) measurements, or the displacements they produced, we propose two real-time instantaneous velocity determination methods based on the moving-window polynomial model. The two methods are called the observation-domain and the coordinate-domain, correspondingly. In the calculation process, the Cholesky update method is utilized to improve numerical efficiency. The proposed methods are a finite impulse in nature, and hence are robust against modeling uncertainties, such as temporal outlying measurements, inaccurate information of noise statistics. Two vehicle-mounted experiments were performed to verify the performance of the coordinate-domain and observation-domain methods. The reference velocities used to evaluate the accuracy of velocity estimates are chosen as GNSS/INS smoothing solution and high-rate RTK solution in these two experiments, respectively. The results show that the coordinate-domain and observation-domain methods were comparable in performance, both better than that of the Nonlinear Tracking Differentiator (NTD) method. Specifically, the improvement in terms of root mean square of the proposed methods, compared with the NTD, can be 29.3-40.3% and 6.25-16.4% in the two experiments, respectively.