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.     

Exponential Stability for Time-changed Stochastic Differential Equations

Acta Mathematicae Applicatae Sinica, English Series - So far there have been few results presented on the exponential stability for time-changed stochastic differential equations. The main aim of...     

Quantum Resources in Heisenberg XX Model Under Noisy Environment

International Journal of Theoretical Physics - In this paper, we investigate Einstein-Podolsky-Rosen steering, Bell non-locality, first-order coherence and concurrence in the spin- $\frac {1}{2}$...     

Boundedness of Singular Integral Operators on Local Hardy Spaces and Dual Spaces

Potential Analysis - The purpose of this paper is to provide necessary and sufficient conditions of the boundedness for singular integrals on the local Hardy space and its dual. Particularly the...     

Interfacial instability of ferrofluid flow under the influence of a vacuum magnetic field

This study is to numerically test the interfacial instability of ferrofluid flow under the presence of a vacuum magnetic field. The ferrofluid parabolized stability equations(PSEs) are derived from the ferrofluid stability equations and the Rosensweig equations, and the characteristic values of the ferrofluid PSEs are given to describe the ellipticity of ferrofluid flow. Three numerical models representing specific cases considering with/without a vacuum magnetic field or viscosity are created to mathematically examine the interfacial instability by the computation of characteristic values. Numerical investigation shows strong dependence of the basic characteristic of ferrofluid Rayleigh-Taylor instability(RTI) on viscosity of ferrofluid and independence of the vacuum magnetic field.For the shock wave striking helium bubble, the magnetic field is not able to trigger the symmetry breaking of bubble but change the speed of the bubble movement. In the process of droplet formation from a submerged orifice, the collision between the droplet and the liquid surface causes symmetry breaking. Both the viscosity and the magnetic field exacerbate symmetry breaking. The computational results agree with the published experimental results.     

Diversity-oriented synthesis and antifungal activities of novel pimprinine derivative bearing a 1,3,4-oxadiazole-5-thioether moiety

Molecular Diversity - Based on the strategy of diversity-oriented synthesis and the structures of natural product pimprinine and streptochlorin, two series of novel pimprinine derivatives...     

Stability and control of radial deployment of electric solar wind sail

Nonlinear Dynamics - The paper studies the stability and control of radial deployment of an electric solar wind sail with the consideration of high-order modes of elastic tethers. The electric...     

Fully Chemical Recyclable Poly(γ-butyrolactone)-based Copolymers with Tunable Structures and Properties

Chinese Journal of Polymer Science - The emerging chemical recyclable polymers, such as poly(γ-butyrolactone) (PGBL) and poly((R)-3,4-trans six-membered ring-fused GBL) (P((R)-M)), provide a...     

Biotemplated Fabrication of ZnO Microstructures with Enhanced Photocatalytic Properties from Rape Pollen

Crystallography Reports - Sodium dodecyl sulfonate and rape pollen were selected as soft and hard templates to induce ZnO formation under hydrothermal conditions, respectively. In the preparation...