Click-through rate is a central issue in ad recommendation and has recently received extensive research attention in academia and industry. Research shows that the accuracy of prediction results in ...CTR prediction is closely related to interactive features and user interest features. However, existing models usually focus on one aspect of features, i.e., interactive features or interest features, and few studies have attempted to learn both interactive features and interest features simultaneously. In this paper, a novel model called CFF as an abbreviation for Combining interactive Features and interest Features is proposed to learn interactive features and user interest features simultaneously. To efficiently learn fine-grained interactive features, an attention-based squeeze equal interaction network (ASENet) is constructed to select salient feature information at the level of equal interactive features. A bi-directional attention-target item gated recurrent unit (Bi-ATGRU) is designed to learn the dependencies between user interests and items. Specifically, it refines and integrates interest features by incorporating context information, historical behaviors, and target item. Extensive experiments on four public datasets indicate CFF outperforms other baselines in terms of evaluation metrics (the Logloss decreases by 1.97% on Frappe and 1.85% on MovieLens).
•Integrated design of material and structure avails the EWAP.•Foam concrete is the optimum cementitious electromagnetic absorption matrix.•Layer structure makes full use of absorbers with different ...absorption mechanisms.•Optimum EWAP with RL < -10 dB in full-band of 2–18 GHz is achieved.
The integrated design of materials and structures is essential for achieving superior electromagnetic wave absorption performance (EWAP) in cementitious materials. In this study, multicomponent microwave absorbents are designed to improve the EWAP of foamed cementitious materials. The effects of the type, content, and single/compound doping of the absorbents and material density on the EWAP of multilayer foamed cementitious materials are tested. The experimental results show that the multilayer structure can enhance the impedance-matching characteristics between cementitious materials and air. It also conduces more electromagnetic waves to penetrate deeper into the cementitious materials to be absorbed. The electromagnetic dissipation of cementitious materials can be significantly improved by combining the effects of interlayer reflection scattering, interface polarization, and resonance. Furthermore, the multilayer structure offers flexibility in adjusting the type and content of the absorbent in each layer and endows each layer with different electromagnetic dissipation mechanisms, resulting in a significantly improved EWAP. For four layers of foamed cementitious materials, the EWAP is optimized using three different iron ores (WMHL-135: 10 wt% magnetite, 30 wt% hematite, and 50 wt% limonite) in the absorbing layers. It achieves the optimum EWAP with a full-band effective absorption bandwidth of reflection loss (RL) < -10 dB, in the frequency range of 2–18 GHz, a minimum RL of −37.0 dB, and an average RL of –23.4 dB.
Multi-label text classification task is one of the research hotspots in the field of natural language processing. However, most of the existing multi-label text classification models are only ...suitable for scenarios with a small number of labels and coarser granularity. Aiming at the problem of difficulty in obtaining sequence information and obvious lack of semantic information when the text sequence grows, this paper proposes an R-Transformer_BiLSTM model based on label embedding and attention mechanism for multi-label text classification. First, we use the R-Transformer model to obtain the global and local information of the text sequence in combination with part-of-speech embedding. At the same time, we use BiLSTM+CRF to obtain the entity information of the text, and use the self-attention mechanism to obtain the keywords of the entity information, and then use bidirectional attention and label embedding to further generate text representation and label representation. Finally, the classifier performs text classification according to the label representation and text representation. In order to evaluate the performance of the model, we conducted a lot of experiments on the RCV1-V2 and AAPD datasets. Experimental results show that the model can effectively improve the efficiency and accuracy of multi-label text classification task.
High levels of apolipoprotein C3 (APOC3) can lead to hypertriglyceridemia, which increases the risk of cardiovascular disease. We aim to create APOC3-knockout (KO) rabbits and explore the effects of ...APOC3 deletion on the occurrence and development of atherosclerosis.
An sgRNA anchored to exon 2 of APOC3 was designed to edit embryo genomes using the CRISPR/Cas9 system. The founder rabbits were sequenced, and their lipid profile, inflammatory cytokines, and atherosclerotic plaques were analyzed.
When given a normal chow (NC) diet, all APOC3-KO rabbits had 50% lower triglyceride (TG) levels than those of the matched age control group. Additionally, their plasma lipoprotein lipase increased. When fed a high-fat diet, APOC3 deficiency was observed to be more conducive to the maintenance of plasma TG, total cholesterol, and low-density lipoprotein cholesterol levels, and the inhibition of the inflammatory response and the protection against atherosclerosis in rabbits.
APOC3 deficiency can delay the formation of atherosclerosis-induced HFD in rabbits, indicating this is a novel therapeutic target to treat atherosclerosis.
Effective separation and transfer of photoexcited carriers are essential for photocatalysis, which could be optimized by the rational design of morphology and phase structure. Herein, using NH4HCO3 ...as intercalating and orienting agents, few-layer vertical-standing multiphasic MoS2 nanosheets (VM-MoS2) were successfully constructed along the longitudinal axis of CdS nanorods via a green hydrothermal method. The growth mechanism of VM-MoS2 on CdS nanorods was demonstrated to include the competitive adsorption of HCO3 – and Mo7O24 6– anions on the protonated CdS, and the intercalation of NH4 + cations. Moreover, the impacts of the morphology and phase structure of the MoS2 cocatalyst on the photocatalytic H2 evolution (PHE) performance of CdS were carefully compared. It is found that the VM-MoS2 nanosheets can not only expose abundant active sites, but also allow the supporter to harvest the light effectively. Additionally, the metallic 1T-MoS2 in VM-MoS2 would help the interface transfer of photogenerated electrons and act as photoelectron entrepots and catalytically active sites for H2 evolution. With the VM-MoS2 cocatalyst, the as-synthesized CdS@VM-MoS2 shows an outstanding PHE rate of ∼40.1 mmol·h–1·g–1 under visible-light irradiation. Interestingly, integrating with the 2H-MoS2 phase will make the metastable 1T-MoS2 more stable, leading to the exceptional photostability of CdS@VM-MoS2 nanocomposite.
Display omitted
•Lignin-based carbon dots is used as the support of Pt single atoms for the first time.•Isolated Pt atoms can bond with the Pyridinic N of LCDs to form a Pt-N4 structure.•The enhanced ...electron transition in Pt-LCDs can make more electrons participate in HER.•Pt-LCDs@CdS shows ultrahigh photocatalytic H2 evolution performance and photostability.
Single-atom metal species always run the risk of aggregating into large particles, thus it is essential to develop an appropriate support for Pt single atoms to prevent their aggregation on the premise of ensuring catalytic activity. Herein, N-doped lignin-based carbon dots (NLCDs) prepared by a facile hydrothermal method is used as the support of Pt single atoms for the first time. Through mild photo-assistant method, isolated Pt atoms can bond with the Pyridinic N of NLCDs to form Pt-N4 coordination structures. Moreover, the enhanced π → π* and n → π* electron transitions in Pt-NLCDs make more electrons participate in H2-evolution. The Pt-NLCDs@CdS ternary photocatalyst shows an exceptional visible-light-driven photocatalytic H2-evolution (PHE) rate up to ∼ 46.10 mmol·h−1·g−1. Notably, the ternary system shows ultrahigh photostability in the 60 h continuous photolysis application. Photoelectrochemical analyses demonstrate the synergetic effect of Pt single atoms, NLCDs and CdS will boost the separation, transport and reaction of photogenerated carriers.
To convert the disadvantages of CdS photocorrosion into the advantages of cocatalysts loading, a novel cocatalyst loading strategy combining photo-deposition and ion-exchange is designed to boost the ...photocatalytic H2 evolution performance of CdS.
Display omitted
•Convert the disadvantages of CdS photocorrosion into the advantages of cocatalysts loading.•Design a novel loading strategy combining photocorrosion-deposition and ion-exchange.•Few-layer vertical-standing Mo-S nanofins can be facile formed on CdS.•The composite shows an ultrahigh-efficient photocatalytic H2 evolution performance.
The promoting role of cocatalysts is not only dependent on their physical and chemical properties, but also closely related to their loading methods and nanostructures. In this study, by converting the disadvantages of CdS photocorrosion into the advantages of cocatalysts loading, a novel cocatalyst loading strategy is adopted to promote the photocatalytic H2 evolution performance of CdS. Using 1D CdS nanorods as primary photocatalyst, a limited amount of Mo-O cocatalyst (e.g. CdMoO4-MoOx) was in-situ deposited on the photocorrosive surface of CdS under intensive irradiation, and further transformed into few-layer Mo-S (e.g. MoS2-CdMoS4) nanofins by a facile ion-exchange method. The as-synthesized CdS@MoS2-CdMoS4 heterojunctions show a surprising “boiling” effect in visible-light-driven H2 evolution process (∼40.3 mmol·h−1·g−1).
Target recovery through scattering media is an important aspect of optical imaging. Although various algorithms combining deep-learning methods for target recovery through scattering media exist, ...they have limitations in terms of robustness and generalization. To address these issues, this study proposes a data-decoupled scattering imaging method based on autocorrelation enhancement. This method constructs basic-element datasets, acquires the speckle images corresponding to these elements, and trains a deep-learning model using the autocorrelation images generated from the elements using speckle autocorrelation as prior physical knowledge to achieve the scattering recovery imaging of targets across data domains. To remove noise terms and enhance the signal-to-noise ratio, a deep-learning model based on the encoder–decoder structure was used to recover a speckle autocorrelation image with a high signal-to-noise ratio. Finally, clarity reconstruction of the target is achieved by applying the traditional phase-recovery algorithm. The results demonstrate that this process improves the peak signal-to-noise ratio of the data from 15 to 37.28 dB and the structural similarity from 0.38 to 0.99, allowing a clear target image to be reconstructed. Meanwhile, supplementary experiments on the robustness and generalization of the method were conducted, and the results prove that it performs well on frosted glass plates with different scattering characteristics.