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.     

Terahertz radiation from pentacene organic diode at room temperature

We investigate terahertz radiation(T-rays) from a pentacene organic diode at room temperature. The quantum chemistry calculation for frequency-related Huang–Rhys factor of pentacene is also carried out. The results demonstrate that the T-rays can come from a bending vibration of pentacene skeleton after the energy of pentacene exciton transferring to the vibrational excited state via electron–phonon coupling. Frequency and natural bond orbital analytics of pentacene and its derivatives are performed in order to explain the result and develop new materials to get higher emission. This work provides a new way to produce T-rays with a simple device at room temperature.     

Selective Catalytic Isomerization of β-Pinene Oxide to Perillyl Alcohol Enhanced by Protic Tetraimidazolium Nitrate

A series of tetraimidazolium salts with different anions was prepared and applied in the isomerization of β-pinene oxide. After examining the activity of different catalysts, a remarkable enhancement of the selectivity of perillyl alcohol (47 %) was obtained over [PEimi][HNO₃]₄ under mild reaction conditions and using DMSO as the solvent. Furthermore, noncovalent interactions between solvent molecules and the catalyst were found by FT-IR spectroscopy and confirmed by computational chemistry. The homogeneous catalyst showed excellent stability and was reused up to six times without significant loss.     

盐酸伊达比星杂质A的合成工艺优化及其高效液相分析

通过对盐酸伊达比星脱羟基杂质A的合成研究，为盐酸伊达比星的质控标准提供参考。以盐酸伊达比星为起始原料，经过4步反应得到脱羟基杂质A。本文提供了盐酸伊达比星杂质A的制备方法以及高效液相色谱分析方法，为杂质含量控制提供了依据。该工艺可以稳定、快速制备高纯度脱羟基杂质A，制备总收率为57.20%，纯度为57.20%，其结构经¹H NMR， ¹³C NMR和MS(ESI)表征。     

紫色球杆菌视紫红质光谱特性的机器学习研究

近年来，机器学习等人工智能技术被应用于蛋白质工程，其在蛋白质结构、功能预测、催化活性等研究中具有独特优势。在未知蛋白质结构的情况下，将蛋白质序列和功能特性与机器学习相结合，基于序列-活性关系（innovative sequence-activity relationship，ISAR）算法，将蛋白质氨基酸序列数字化，用快速傅里叶变换（fast four transform，FFT）进行预处理，再进行偏最小二乘回归建模，可在数据集较少情况下拟合得到最佳模型。通过机器学习对紫色球杆菌视紫红质（gloeobacter violaceus rhodopsin，GR）的突变体蛋白质氨基酸序列与光谱最大吸收波长进行建模，获得了最佳模型。用最佳索引LEVM760106建模得到的确定系数R² 为0.944，均方误差E为11.64。用小波变换进行的预处理，其R² 虽也约为0.944，但E大于11.64，不及FFT进行的预处理。方法较好地解决了蛋白质序列与功能特性之间的数学建模问题，在蛋白质工程中可为预测更优的突变体提供支持。     

An experimental study on the fire characteristics of new and aged building wires using a cone calorimeter

Journal of Thermal Analysis and Calorimetry - The fire behavior of one flame-retardant PVC cable used as building wire is investigated in this work. Bench scale tests were performed using a cone...     

Effect of palladium oxide electrode on potentiometric sensor response to carbon monoxide

Ionics - This study aims to explore the potential merits of palladium oxide sensing electrode. PdO is applied on a solid-state potentiometric sensor on an yttria-stabilized zirconia (YSZ)...