基于GA_XGBoost模型的大学生科研能力预测问题研究

科学评价大学生科研创新能力对我国科研水平的提高具有重要意义.采用机器学习模型来预测大学生科研能力可以起到良好的效果,提出一种GAXGBoost模型来实现对大学生的科研能力预测.此模型是以Xgboost算法为基础,然后充分利用遗传算法的全局搜索能力自动搜索Xgboost最优超参数,避免了人为经验调参不准确的缺陷,最后采用精英选择策略以此确保每一轮都是最佳的进化结果.通过分析表明,所采用的GAXGBoost模型在大学生科研能力预测的结果中具有很高的精度,将此模型与Logistic Regression、Random Forest、SVM等模型进行对比,GAXGBoost模型的预测精度最高.     

Equation for Condensed Matter Phase Transitions

Russian Physics Journal - A unified equation for the pressure drop in the apparatus with the stationary and fluidized granular layers is derived. The resulting recurrent equation is used to...     

The Interference of Magnetoelastic Waves in an Yttrium Iron Garnet Film

Physics of the Solid State - The interference of excited and received by current antennae magnetoelastic waves counterpropagating in an yttrium iron garnet film is experimentally studied. It is...     

Developing Flexible Quinacridone-Derivatives-Based Photothermal Evaporaters for Solar Steam and Thermoelectric Power Generation

Solar-driven interfacial vaporization by localizing solar-thermal energy conversion to the air−water interface has attracted tremendous attention. In the process of converting solar energy into heat energy, photothermal materials play an essential role. Herein, a flexible solar-thermal material di-cyan substituted 5,12-dibutylquinacridone (DCN−4CQA)@Paper was developed by coating photothermal quinacridone derivatives on the cellulose paper. The DCN−4CQA@Paper combines desired chemical and physical properties, broadband light-absorbing, and shape-conforming abilities that render efficient photothermic vaporization. Notably, synergetic coupling of solar-steam and solar-electricity technologies by integrating DCN−4CQA@Paper and the thermoelectric devices is realized without trade-offs, highlighting the practical consideration toward more impactful solar heat exploitation. Such solar distillation and low-grade heat-to-electricity generation functions can provide potential opportunities for fresh water and electricity supply in off-grid or remote areas.     

On a Lemma Arising in the Solution of Waring’s Problem

Siberian Mathematical Journal - We give a simple proof of one assertion used in solving Waring’s problem.     

Alkaline oxygen evolution: exploring synergy between fcc and hcp cobalt nanoparticles entrapped in N-doped graphene

Herein, we report a Mott-Schottky catalyst by entrapping cobalt nanoparticles inside the N-doped graphene shell (Co@NC). The Co@NC delivered excellent oxygen evolution activity with an overpotential of merely 248 mV at a current density of 10 mA cm^–2 with promising long-term stability. The importance of Co encapsulated in NC has further been demonstrated by synthesizing Co nanoparticles without NC shell. The synergy between the hexagonal close-packed (hcp) and face-centered cubic (fcc) Co plays a major role to improve the OER activity, whereas the NC shell optimizes the electronic structure, improves the electron conductivity, and offers a large number of active sites in Co@NC. The density functional theory calculations have revealed that the hcp Co has a dominant role in the surface reaction of electrocatalytic oxygen evolution, whereas the fcc phase induces the built-in electric field at the interfaces with N-doped graphene to accelerate the H⁺ ion transport.     

Fully asynchronous stochastic coordinate descent: a tight lower bound on the parallelism achieving linear speedup

Mathematical Programming - We seek tight bounds on the viable parallelism in asynchronous implementations of coordinate descent that achieves linear speedup. We focus on asynchronous coordinate...     

Optimal design of gears contact interface modification for an objective as minimum impact resistance of initial meshing-in time domain

Meccanica - In gear pair actual alternating meshing process, the comprehensive errors of the transmission system and the thermal elastic deformation of the teeth body cause the gears in the meshing...