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.     

Setting local rank constraints by orthogonal projections for image resolution analysis: Application to the determination of a low dose pharmaceutical compound

Raman chemical imaging provides chemical and spatial information about pharmaceutical drug product. By using resolution methods on acquired spectra, the objective is to calculate pure spectra and distribution maps of image compounds. With multivariate curve resolution-alternating least squares, constraints are used to improve the performance of the resolution and to decrease the ambiguity linked to the final solution. Non negativity and spatial local rank constraints have been identified as the most powerful constraints to be used.     

The Effect of the Spacer of Bis(biurea) Ligands on the Structure of A2L3‐type (A=anion) Phosphate Complexes

By tuning the length and rigidity of the spacer of bis(biurea) ligands L, three structural motifs of the A₂L₃ complexes (A represents anion, here orthophosphate PO₄^3?), namely helicate, mesocate, and mono‐bridged motif, have been assembled by coordination of the ligand to phosphate anion. Crystal structure analysis indicated that in the three complexes, each of the phosphate ions is coordinated by twelve hydrogen bonds from six surrounding urea groups. The anion coordination properties in solution have also been studied. The results further demonstrate the coordination behavior of phosphate ion, which shows strong tendency for coordination saturation and geometrical preference, thus allowing for the assembly of novel anion coordination‐based structures as in transition‐metal complexes.     

Immobilizing redox enzymes at mesoporous and nanostructured electrodes

Three key challenges are stimulating intensive research in the development of productive direct electron transfer mode enzyme electrodes: proper enzyme orientation, high enzyme loading, and full retention of enzyme activity. In this review, we summarize some significant advances that have been reported in the last years on the design of mesoporous and nanostructured electrodes as enzyme scaffolds and of innovative methodologies for wiring enzymes to electrodes. Particular attention is given to investigations on physical factors that determine a favorable enzyme immobilization, to provide rational guidelines for the design of productive enzymatic electrodes. Finally, some emerging trends focused on the spatial organization of either single enzymes or enzyme cascades are also briefly addressed.     

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.     

Identification of metabolic markers in patients with type 2 Diabetes by Ultrafast gas chromatography coupled to electronic nose. A pilot study

Metabolomics is a potential tool for the discovery of new biomarkers in the early diagnosis of diseases. An ultra-fast gas chromatography system equipped to an electronic nose detector (FGC eNose) was used to identify the metabolomic profile of Volatile Organic Compounds (VOCs) in type 2 diabetes (T2D) urine from Mexican population. A cross-sectional, comparative, and clinical study with translational approach was performed. We recruited twenty T2D patients and twenty-one healthy subjects. Urine samples were taken and analyzed by FGC eNose. Eighty-eight compounds were identified through Kovats's indexes. A natural variation of 30% between the metabolites, expressed by study groups, was observed in Principal Component 1 and 2 with a significant difference (p < 0.001). The model, performed through a Canonical Analysis of Principal coordinated (CAP), allowed a correct classification of 84.6% between healthy and T2D patients, with a 15.4% error. The metabolites 2-propenal, 2-propanol, butane- 2,3-dione and 2-methylpropanal, were increased in patients with T2D, and they were strongly correlated with discrimination between clinically healthy people and T2D patients. This study identified metabolites in urine through FGC eNose that can be used as biomarkers in the identification of T2D patients. However, more studies are needed for its implementation in clinical practice.     

基于深度学习的ELM实时识别研究

基于深度学习的方法，在HL-2A装置上开发出了一套边缘局域模(ELM)实时识别算法。算法使用5200次放电数据(约24.19万数据切片)进行学习，得到一个深度为22层的卷积神经网络。为衡量算法的识别能力，识别了HL-2A装置自2009年实现稳定ELMy H模放电以来所有历史数据(约26000次放电数据)，共识别出1665次H模放电，其中误识别35次，误报率为2.10%。在实际的1634次H模放电中，漏识别4次，漏识别率为0.24%。该误报率和漏报率可以满足ELM实时识别的精度要求。识别算法在实时控制环境下，对单个时间点的平均计算时间为0.46ms，可以满足实时控制的计算速度要求。