Search

Session-based recommendation aims to predict user preferences based on anonymous behavior sequences. Recent research on session-based recommendation systems has mainly focused on utilizing attention mechanisms on sequential patterns, which has achieved significant results. However, most existing studies only consider individual items in a session and do not extract information from continuous items, which can easily lead to the loss of information on item transition relationships. Therefore, this paper proposes a session-based recommendation algorithm (SGT) based on Gated Recurrent Unit (GRU) and Transformer, which captures user interests by learning continuous items in the current session and utilizes all item transitions on sessions in a more refined way. By combining short-term sessions and long-term behavior, user dynamic preferences are captured. Extensive experiments were conducted on three session-based recommendation datasets, and compared to the baseline methods, both the recall rate Recall@20 and the mean reciprocal rank MRR@20 of the SGT algorithm were improved, demonstrating the effectiveness of the SGT method.

Precipitation is a significant index to measure the degree of drought and flood in a region, which directly reflects the local natural changes and ecological environment. It is very important to grasp the change characteristics and law of precipitation accurately for effectively reducing disaster loss and maintaining the stable development of a social economy. In order to accurately predict precipitation, a new precipitation prediction model based on extreme learning machine ensemble (ELME) is proposed. The integrated model is based on the extreme learning machine (ELM) with different kernel functions and supporting parameters, and the submodel with the minimum root mean square error (RMSE) is found to fit the test data. Due to the complex mechanism and factors affecting precipitation change, the data have strong uncertainty and significant nonlinear variation characteristics. The mean generating function (MGF) is used to generate the continuation factor matrix, and the principal component analysis technique is employed to reduce the dimension of the continuation matrix, and the effective data features are extracted. Finally, the ELME prediction model is established by using the precipitation data of Liuzhou city from 1951 to 2021 in June, July and August, and a comparative experiment is carried out by using ELM, long-term and short-term memory neural network (LSTM) and back propagation neural network based on genetic algorithm (GA-BP). The experimental results show that the prediction accuracy of the proposed method is significantly higher than that of other models, and it has high stability and reliability, which provides a reliable method for precipitation prediction.

In this work, we predict two novel two-dimensional (2D) auxetic materials, BiSbX3 (X = S, Se) monolayers, through first-principles calculations. Attributed to their special braceletlike structure, the in-plane negative Poisson's ratio (NPR) of BiSbS3 and BiSbSe3 monolayers are as high as -0.25 and -0.26, respectively. The phonon dispersion calculations, ab initio molecular dynamics simulations, and elastic constants calculations demonstrate that these two monolayers possess excellent dynamic, thermal, and mechanical stabilities. The band gap values of BiSbS3 and BiSbSe3 calculated at the HSE level by considering the spin-orbit coupling (SOC) effect are 1.68 and 1.20 eV. The anisotropic carrier mobility and superior optical absorption indicate that they may shine in the next generation of electronic and optoelectronic devices. All of these discoveries not only enrich the types of auxetic materials but also provide a structural reference for designing new auxetic materials on the molecular level. Furthermore, they can provide theoretical guidance for future applications of BiSbX3 (X = S, Se) monolayers in various fields.

The gas-sensing performance of PbSnS2 monolayer to sulfur hexafluoride (SF6) decompositions (SO2, H2S, HF, SOF2, and SO2F2) has been systematically studied by density functional theory and nonequilibrium Green’s function calculations. The most stable adsorption configurations, adsorption energies, charge transfer, electron localization function (ELF), density of states (DOS), and ${I}$ – ${V}$ curves of gas molecules on PbSnS2 are calculated and discussed. SO2 exhibits the largest adsorption energy of −0.74 eV with the charge transfer of −0.39 e and HF possesses the biggest charge transfer of −0.66 e with the adsorption energy of −0.46 eV. The results of ELF and DOS show that the interaction between SO2 and PbSnS2 may be stronger than those of other gases. Moreover, ${I}$ – ${V}$ curves show that the responses of PbSnS2 to SO2 is up to 65.5% at the bias of 2.2 V. Therefore, PbSnS2 monolayer may be a qualified material for detecting SF6 decompositions.

In this work, we explore the application potential of AsP/M2CO2 (M = Sc, Zr) van der Waals heterostructures in photocatalytic water splitting through the first-principles calculations. The calculated results show that AsP/Zr2CO2 heterostructure possesses an unfavorable type-Ⅰ band alignment, whereas AsP/Sc2CO2 exhibits a desirable type-Ⅱ band alignment, which is beneficial for separating the photogenerated electron-hole pairs. Also, the band edge positions of AsP/Sc2CO2 heterostructure stride the redox potential of water, ensuring favorable reaction kinetics. Besides, the strong optical absorption of AsP/Sc2CO2 heterostructure in both visible and ultraviolet regions (especially up to 10−6 cm−1 at about 250 nm) makes it possible to utilize solar energy effectively. Meanwhile, AsP/Sc2CO2 heterostructure has an exciton binding energy as low as 0.09 eV, which quantitatively illustrates the high separation efficiency of photogenerated charge carrier. Thus, the type-Ⅱ band alignment, suitable band edge position, strong light absorption, and low exciton binding energy together indicate that AsP/Sc2CO2 heterostructure is a potential photocatalytic material. In addition, the obvious redshift phenomenon in the optical spectrum of AsP/Sc2CO2 heterostructure shows that biaxial strain can improve its light capture capability. Also, the interconversion between type-Ⅱ and type-Ⅰ can be achieved by applying different strains. All these findings suggest that the novel AsP/Sc2CO2 heterostructure has significant application prospects in next-generation photovoltaic and photocatalytic devices.

Two-dimensional Janus nanomaterials, due to their unique electronic, optical, and piezoelectric characteristics resulting from the antisymmetric structures, exhibit great prospects in multifunctional energy application to alleviate the energy crisis. Monolayer Janus Pb2SSe, with a black phosphorus-like structure and an indirect band gap of 1.59 eV as well as high carrier mobility (526-2105 cm2 V-1 s-1), displays outstanding potentials in the energy conversion between nanomechanical energy, solar energy, waste heat, and electricity, which has been comprehensively studied utilizing DFT-based simulations. The research results reveal that monolayer Pb2SSe not only possesses giant in-plane piezoelectricity of d11 = 75.1 pm V-1 but also superhigh out-of-plane piezoelectric coefficients (d31 = 0.5 pm V-1 and d33 = 15.7 pm V-1). Meanwhile, by constructing Pb2SSe bilayers, the out-of-plane piezoelectric coefficients can be significantly enhanced (d31 = 19.2 pm V-1 and d33 = 194.7 pm V-1). In addition, owing to the small conduction band offset, suitable donor band gap and excellent light absorption capability in the Pb2SSe/SnSe (Pb2SSe/GeSe) heterostructure, the power conversion efficiencies were calculated to be up to 20.02% (Pb2SSe/SnSe) and 19.28% (Pb2SSe/GeSe), making it a promising candidate for solar energy collection. Furthermore, from the thermoelectric electron and phonon transport calculations, it can be found that the Pb2SSe monolayer is an n-type thermoelectric material with ultrahigh ZT = 2.19 (1.52) at room temperature, which can be traced back to its ultralow κL = 0.78 (0.99) W m-1 K-1, and superhigh PF = 10.18 (8.25) mW m-1 K-2 along the x(y) direction at the optimal doping concentration at 300 K. The abovementioned versatile characteristics in the Janus Pb2SSe monolayer, along with its comprehensive stabilities (energy, dynamic, thermal, and mechanical stabilities), highlight its potential in clean energy harvesting.

In this work, we predict two novel two-dimensional (2D) auxetic materials, BiSbX

Two-dimensional Janus nanomaterials, due to their unique electronic, optical, and piezoelectric characteristics resulting from the antisymmetric structures, exhibit great prospects in multifunctional energy application to alleviate the energy crisis. Monolayer Janus Pb

Thermal transfer enhanced multifunctional energy conversion form-stabilized composite phase change materials (PCMs) present attractive research prospects for the energy management materials. However, complicated preparation, unsatisfied heat storage capacity and limited thermal transfer enhancement restrict their application. In this work, high-performance thermal transfer enhanced composited PCMs are facilely prepared via integration of polypyrrole/Fe3O4- functionalized hollow kapok fiber (KF) aerogel supports with paraffin wax (PW). The prepared composite PCMs achieve high thermal storage density (melting enthalpy of 161.4 J/g for the PW load mass fractions of 88%), greatly enhancement of thermal conductivity (1.06 W/m K, 307% higher than pristine PW) and outstanding cycling thermal stability. The hollow kapok fiber aerogel supports prevent the leakage and improve loading amount, and also endow the renewable and pollution-free features for the composite PCMs, which provides great potential for the commercial application on a large scale. Simultaneously, the continuous polypyrrole (PPy) dual coating and Fe3O4 nanoparticles endow the composite PCMs not only with excellent thermal conductivity, but also with solar /magnetic- thermal energy conversion characteristics which enriches the energy storage source of the materials.

Cu paste has been considered as the most promising die-attach material for high-power devices. To investigate the influences of Cu paste layer thickness on bonding qualities of joint, the large-area joints (10 × 10 mm2) were bonded by multiscale Cu paste with a layer thickness varying from 30 to 200 μm, and the mechanical properties and microstructure characteristics of joints were analyzed. A layer thickness between 80 and 100 μm is considered the optimal thickness, in which, the shear strength of joint reaches the maximum value of 26.1Mpa and the bond layer is the most tightly connected. We believe that a suitable thickness not only ensures the complete sintered of the Cu particles, but also inhibits the coursing on the pores, which enhance the joint bonding strength. The results of this work help to determine the thickness of paste and are useful for power electronics packaging.

We aimed to investigate the risk factors for cervical intraepithelial neoplasia (CIN) in Jiexiu, Shanxi Province, China. Twenty thousand eligible married women (age: 18-65 years) were administered with a questionnaire on potential risk factors for CIN and underwent liquid based Pap test. All women with abnormal cytological results underwent colposcopy with biopsy. Based on the biopsy pathology results, women were then assigned to either study group (with CIN) or control group (negative for histological results and volunteered to participate in the follow up study). The women in both study group and control group underwent vaginal microflora detection and dietary survey. The potential risk factors were analyzed by using ordinal logistic regression. Among the 20,000 women ne 1,438 women (7.19%) had cytologic abnormalities and 410 (2.05%) women were diagnosed histologically with CIN lesions, including 317 (1.58%) with CIN1, 93 (0.50%) with CIN2/3and 11 (55/100,000) with squamous cell carcinoma (SCC). The average daily dietary folate intake was significantly lower in the study group (344.61±153.07μg) than in the control group (371.50±166.58μg; P

Sintered Cu is considered as the most promising strategy in die-attachment, since its capacities of low-temperature bonding and high-temperature working. Many studies have reported the excellent properties of sintered Cu, but there is still some room for improvement. Here, strong Cu–Cu joints and insulated-gate bipolar transistor (IGBT) devices are achieved by sintering Cu paste and three composite pastes, and the mechanical, electrical, and microstructure properties of joints are investigated. The shear strength of pure Cu bonded joint is 23.4 MPa, which is improved to 26.64 MPa by introducing 0.6 wt% N-doped multi-walled carbon nanotubes (N-doped MWCNTs) into Cu paste. Excellent electrical properties are achieved as the resistivities of sintered Cu/N-doped MWCNTs layer and Cu/carboxylated WMCNTs are 2.252 and 2.551 μ Ω·cm, respectively, which are lower than the sintered pure Cu layer (3.473 μ Ω·cm). In addition, the functionalized MWCNTs are found exerted positive effects on the joint reliability. The enhancements on the joint qualities are attributed to that the functionalized MWCNTs improve the affinity of the Cu and CNTs and increase the density of bonding layer. This article provides an effective way to improve the sintered Cu joints sintering qualities, which is suitable for the attachment of next-generation devices and substrates.

In this work, recovery performance of a novel sort of form stable phase change materials (FSPCMs) with simultaneously shape memory properties was investigated systematically. The FSPCMs comprising hexadecane as latent heat storage material and olefin block copolymer (OBC) as supporting material were prepared by simple swelling preparation method. This kind of difunctional materials exhibits fantastic shape memory properties with mass fraction of hexadecane reaching 70 wt%, which indicates excellent energy storage ability as well. It is found that the recovery performance under 50% strain is obviously better than that under higher strains for all the specimens investigated, while surprisingly, the composite with 70 wt% hexadecane exhibits better recovery performance than pure OBC as well as the composite with 30 wt% hexadecane at strains of 100 and 200%. The structural change brought by incorporation of hexadecane was explored with in situ two-dimensional X-ray scattering measurements. According to experiment results, we find that the recovery performance is closely correlated to change of long period, indicating the irreversible deformation of the difunctional materials may originate from lamellar fragmentation of OBC, while addition of massive hexadecane is beneficial to restrain lamellar fragmentation and thus improves the recovery performance. We believe this exploration may shed light on comprehending the relationship between structural change and shape memory performance, thus contributing to adjusting the properties of the difunctional materials and expanding their potential applications.