Factorization Machines (FMs) are effective in incorporating side information to overcome the cold-start and data sparsity problems in recommender systems. Traditional FMs adopt the inner product to model the second-order interactions between different attributes, which are represented via feature vectors. The problem is that the inner product violates the triangle inequality property of feature vectors. As a result, it cannot well capture fine-grained attribute interactions, resulting in sub-optimal performance. Recently, the Euclidean distance is exploited in FMs to replace the inner product and has delivered better performance. However, previous FM methods including the ones equipped with the Euclidean distance all focus on the attribute-level interaction modeling, ignoring the critical intrinsic feature correlations inside attributes. Thereby, they fail to model the complex and rich interactions exhibited in the real-world data. To tackle this problem, in this paper, we propose a FM framework equipped with generalized metric learning techniques to better capture these feature correlations. In particular, based on this framework, we present a Mahalanobis distance and a deep neural network (DNN) methods, which can effectively model the linear and non-linear correlations between features, respectively. Besides, we design an efficient approach for simplifying the model functions. Experiments on several benchmark datasets demonstrate that our proposed framework outperforms several state-of-the-art baselines by a large margin. Moreover, we collect a new large-scale dataset on second-hand trading to justify the effectiveness of our method over cold-start and data sparsity problems in recommender systems.

In recent decades, white light‐emitting diodes (WLEDs) assembled by blue light‐emitting diodes (LEDs) and down‐conversion phosphors have gained rapid development. Thereinto, red phosphors, especially for the ecofriendly and cost‐effective ones, serve an important role for the preparation of superior WLEDs. Herein, a halide compound with a 0D crystal structure, namely [PPh4]2[Bi2Br8(bp4do)(DMSO)2] (BiBrON, PPh4 = tetraphenylphosphonium, bp4do = 4,4′‐bipyridine 1,1′‐dioxide, DMSO = dimethyl sulfoxide) is synthesized at gram scale via a mechanochemical synthesis method. Photophysical properties of BiBrON are investigated in detail. BiBrON has the highest efficiency (78.20%) of red phosphorescence among the Bi‐based halides. The red emission of BiBrON is blue‐light‐excitable with 99.9% color purity and outstanding thermal stability below 400 K. The optical absorption‐emission mechanism in BiBrON is elucidated by the combination study of experiments and density functional theory calculations. BiBrON can be applied as a red phosphor in the blue LED chip‐based WLED with high color rendering index of 87.8, that is higher than the one fabricated utilizing commercial red phosphor (610 nm), and ideal Commission International de l'Eclairage chromaticity coordinate of (0.3282, 0.3311). This work demonstrates for the first time the great potential of Bi‐inorganic–organic hybrid material as phosphors applied for WLED.

No abstract

No abstract

No abstract