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.     

Mononuclear and Dinuclear Copper Complexes of Tridentate Redox-active Ligands with Tunable H-bonding Donors: Structure,Spectroscopy and H+/e− Reactivity

In this research article, we describe the synthesis and characterization of mononuclear and dinuclear Cu complexes bound by a family of tridentate redox-active ligands with tunable H-bonding donors. The mononuclear Cu-anion complexes were oxidized to the corresponding “high-valent” intermediates by oxidation of the redox-active ligand. These species were capable of oxidizing phenols with weak O−H bonds via H-atom abstraction. Thermodynamic analysis of the H-atom abstractions, which included reduction potential measurements, pK_a determination and kinetic studies, revealed that modification of the anion coordinated to the Cu and changes in the H-bonding donor did not lead to major differences in the reactivity of the “high-valent” CuY complexes (Y: hydroxide, phenolate and acetate), which indicated that the tridentate ligand scaffold acts as the H⁺ and e⁻ acceptor.     

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。     

激光能量对反钙钛矿GaCMn_3薄膜结构与物性的影响

本文采用脉冲激光沉积方法在LaAlO3(001)单晶衬底上制备了反钙钛矿GaCMn3薄膜,通过控制制备过程中脉冲激光的能量,研究了不同激光能量条件对GaCMn3薄膜结构与物理性能的影响.分别利用X射线衍射仪、原子力显微镜、超导量子干涉仪和物理性能测试系统,对所制备的薄膜的晶体结构、表面形貌和磁性、电输运性质进行了研究.结果表明,制备的样品均为具有多个晶面取向的反钙钛矿薄膜,且薄膜结构和物性明显随制备激光能量的变化而变化.当激光能量为450mJ时,制备的薄膜多晶面取向性最弱,结晶性和表面形貌最优良.实验所得的薄膜均表现出顺磁-铁磁-反铁磁相转变,然而转变过程比块材较平缓,同时薄膜的电阻率并未表现出块材中的突变特征,我们推测该现象很可能是由衬底的应力及衬底的晶格膨胀对薄膜反常晶格变化的抑制作用造成的.     

Optic Fiber Interferometer for Absolute Distance Measurement

1.IntroductionAbsolutedistancemeasurementisfarfromanewtopic.However,itisstillafieldstimulatinggreatinterestsnowadaysduetoitSimportantroleinmanufacturingandassembly['J.SincethegreatsuccessachievedbyMichelsonandBenoitwhentheyfirstdevelopedaninterferometertodeterminethestandardmeterintermsofthemonochromaticredcadmiumline,theopticinterferometerhasbeenprovedtobeoneofthemostpreciseandefficientwayindisplacementmeasurementbecauseofitshighdiscriminationandsimplestructure.However,thetraditionalinterfe…     

Global Chaos Synchronization between Two New Different Chaotic Systems via Active Control

We present chaos synchronization between two new different chaotic systems by using active control. The proposed controller ensures that the states of the controlled chaotic response system asymptotically synchronizes the states of the drive system. Numerical simulations are shown to verify the result.     

O′-β′复相Sialon陶瓷中相组成的控制

用Si₃N₄,SiO₂和Al₂O_?作原料,以Y₂O_?为添加剂,研究了O’-β’所在的四个相容性区中组成,于1400—1800℃之间,O’和β’的形成动力学。O’于1400℃开始生成,1600℃达到最大形成量。高于1600℃,O’减少乃至消失。研究发现,O’的这种不稳定性,并非自身热分解的结果。Al含量的增加,液相出现和β’的形成,才是导致O’-Sialon溶解而消失的原因。 此外,认为Si₃N₄-O’(x=0.3)-Y₂Si₂O₇-YAG相容性区,是在制备具有合理的O’-β’平衡相组成的复相陶瓷时,可供实际选用的组成区。     

《光谱学与光谱分析》2007年征订启事

The authors achieved the temporal coherent control in an Er^3＋ doped telluride glass, one kind of disordered solids. The upconversion at 670 nm was modulated and the dephasing time was simulated as 300 fs. Moreover, the photon echo signal gave the dephasing time due to the phonon interaction. The agreement between the two data indicates that the interaction between the wavepacket and the phonon leads to the fast disappearance of wavepackets interference, which is helpful for the tech- nique to be applied to the disordered solids.