应用支持向量机评价太湖富营养化状态

根据湖库富营养化程度评价标准,随机生成大量学习样本,运用支持向量机（SVM）算法建立富营养化评价模型。采用太湖2012年7—9月的监测数据,分别应用SVM模型和线性插值评分法(SCO)对太湖33个监测点的富营养化状况进行了评价。结果表明太湖在2012年7—9月共出现3种营养类型,其中,中营养主要分布于东部湖区,轻度富营养主要分布于湖心区和东部沿岸区,中度富营养主要分布于西北部湖区,太湖水体整体表现为轻度到中度富营养水平。通过SVM和SCO结果的对比分析,发现两种评价方法的结果一致率为78.8%,出现不一致的个例均属于相邻营养等级,表明该SVM模型是有效的,能够应用于太湖的富营养化评价,且具有更好的收敛性和泛化性。     

An integrated approach for assessing the bioreceptivity of glazed tiles to phototrophic microorganisms

A laboratory-based methodology was designed to assess the bioreceptivity of glazed tiles. The experimental set-up consisted of multiple steps: manufacturing of pristine and artificially aged glazed tiles, enrichment of phototrophic microorganisms, inoculation of phototrophs on glazed tiles, incubation under optimal conditions and quantification of biomass. In addition, tile intrinsic properties were assessed to determine which material properties contributed to tile bioreceptivity. Biofilm growth and biomass were appraised by digital image analysis, colorimetry and chlorophyll a analysis. SEM, micro-Raman and micro-particle induced X-ray emission analyses were carried out to investigate the biodeteriorating potential of phototrophic microorganisms on the glazed tiles. This practical and multidisciplinary approach showed that the accelerated colonization conditions allowed different types of tile bioreceptivity to be distinguished and to be related to precise characteristics of the material. Aged tiles showed higher bioreceptivity than pristine tiles due to their higher capillarity and permeability. Moreover, biophysical deterioration caused by chasmoendolithic growth was observed on colonized tile surfaces.     

Effectiveness of Biosorption-Assisted Microfiltration Process for Treatment of Domestic Wastewater

The efficiency of a biosorbent prepared from Eichhornia crassipes roots (ECR) was explored for the treatment of domestic sewage water in combination with low-cost ceramic microfiltration membrane. Batch sorption studies were conducted as a function of biosorbent dose, initial chemical oxygen demand (COD) loading, and temperature. Sorption equilibrium data of varying initial COD values (116–800 mg/L) indicated high potential of ECR for COD removal. Using 0.25 g/L of biosorbent dose, the equilibrium adsorption capacity was obtained as 2480 mg/g at 20°C for an initial COD loading of 800 mg/L. Microfiltration study was performed using ceramic membrane made from composition of α-alumina and clay. The effect of operating parameters on filtration characteristics was observed in terms of permeate flux. Permeate samples were characterized in terms of various parameters both for the direct filtration, as well as biosorbent-assisted filtration. The filtration behavior of wastewater at varying transmembrane pressure was explained using various membrane fouling models. The results suggested that microfiltration of domestic wastewater with incorporation of biosorbent (0.25 g/L) was highly effective for removal of organic load (>90%), turbidity (>99%), and total suspended solids (TSS) (93–95%) and the treated water quality was suitable for reuse in various purposes, such as gardening, floor and car washing, etc.     

The Rights and Blights of the Politics in Indigenous Higher Education

The Seventh Berndt Foundation Biennial LectureIt is clear from their voluminous scholarly output over five decades that Ronald and Catherine Berndt were keen observers of ‘the winds of change’ that were blowing strongly across most of Aboriginal Australia in that period. We contemporary scholars are closely watching both changes emanating from outside our communities and those that we as Indigenous Australians have instigated and moulded from within. A particular focus of this paper is how Indigenous Higher Education policies, programs and practices have been shaped by the politics of Indigenous self-determination and supported by social justice and human rights agendas. These frameworks have also guided argument, advocacy, and activism in the wider arena of Indigenous affairs, providing much of its language and forms of analysis. Today, however, there is evidence of a growing frustration that such frameworks have prevented alternative arguments, priorities, and possibilities from entering public discourse. In this paper, I discuss how these frameworks mutually constitute and uphold forms of analysis in Indigenous academia and Indigenous affairs in ways that fail to engage with the complexity of the contemporary space in which we as Indigenous people find ourselves today.     

Using the concept of pseudo amino acid composition to predict resistance gene against Xanthomonas oryzae pv. oryzae in rice: an approach from chaos games representation

To evaluate the possibility of an unknown protein to be a resistant gene against Xanthomonas oryzae pv. oryzae, a different mode of pseudo amino acid composition (PseAAC) is proposed to formulate the protein samples by integrating the amino acid composition, as well as the Chaos games representation (CGR) method. Some numerical comparisons of triangle, quadrangle and 12-vertex polygon CGR are carried to evaluate the efficiency of using these fractal figures in classifiers. The numerical results show that among the three polygon methods, triangle method owns a good fractal visualization and performs the best in the classifier construction. By using triangle + 12-vertex polygon CGR as the mathematical feature, the classifier achieves 98.13% in Jackknife test and MCC achieves 0.8462.     

基于多特征融合的多分类运动想象脑电信号识别研究

针对目前多分类运动想象脑电识别存在特征提取单一、分类准确率低等问题,提出一种多特征融合的四分类运动想象脑电识别方法来提高识别率。对预处理后的脑电信号分别使用希尔伯特-黄变换、一对多共空间模式、近似熵、模糊熵、样本熵提取结合时频—空域—非线性动力学的初始特征向量,用主成分分析降维,最后使用粒子群优化支持向量机分类。该算法通过对国际标准数据集BCI2005 Data set IIIa中的k3b受试者数据经MATLAB仿真处理后获得93.30%的识别率,均高于单一特征和其它组合特征下的识别率。分别对四名实验者实验采集运动想象脑电数据,使用本研究提出的方法处理获得了72.96%的平均识别率。结果表明多特征融合的特征提取方法能更好的表征运动想象脑电信号,使用粒子群支持向量机可取得较高的识别准确率,为人脑的认知活动提供了一种新的识别方法。     

基于支持向量回归的酚类化合物QSAR建模

目的：采用定量构效关系（QSAR）方法探索酚类化合物的毒性与分子结构参数的关系。方法：基于支持向量回归（SVR）、依均方误差最小原则选择最优核函数,对酚类化合物及其衍生物进行了QSAR研究。结果：不同数据集选取的最优核函数有异,对小样本、非线性等问题,SVR具有较优的稳定性及预测能力,在酚类化合物及其衍生物的QSAR研究中得到了优于原文献方法的独立预测结果。结论：SVR模型具有较好的预测能力,在QSAR及相关研究中可得到更广泛应用。     

Simultaneous removal of chemical oxygen demand (COD), phosphate, nitrate and H2S in the synthetic sewage wastewater using porous ceramic immobilized photosynthetic bacteria

Simultaneous aerobic treatment of COD, phosphate, nitrate and H₂S in a synthetic sewage wastewater was carried out using porous ceramic immobilized photosynthetic bacteria, Rhodobacter sphaeroidesS, Rb. sphaeroidesNR-3 and Rhodopseudomonas palustris. In the batch treatment, effective simultaneous removal of COD (89%), phosphate (77%), nitrate (99%) and H₂S (99.8%) was observed after 48 h. In semi-continuous treatments with dilution rates of 0.17 to 0.75 day^–1under aerobic conditions, simultaneous removal of these four components was also observed after about one month.     

Real-Time Classification of EMG Myo Armband Data Using Support Vector Machine

ObjectivesThis study investigates the performance of the Support Vector Machine (SVM) to classify non-real-time and real-time EMG signals. The study also compares training performance using personalized and generalized data from all subjects. Thus, an idea about the data sets to be used in the training of the real-time classification model has been put forward. In addition, real-time classification results were obtained for ten days, and it was observed how training oneself would affect the classification results.Material and methods:EMG data were acquired for 7 hand gestures from 8 healthy subjects to create the data set: fist, fingers spread, wave-in, wave-out, pronation, supination, and rest. Subjects repeated each gesture 30 times. The Myo armband with 8 dry surface electrodes was used for data acquisition.Results14 features of the EMG signals have been extracted and non-real-time classification has been made for each feature; the highest accuracy of 96.38% was obtained using root mean square (RMS) and integrated EMG features. Three (3) kernel functions of SVM were tested in non-real-time classification and the highest accuracy was obtained with Cubic SVM using 3rd order polynomial. For this reason, Cubic SVM was used for real-time classification using the features that gave the best results in non-real-time classification. A subject repeated the gestures and real-time classification was performed. The highest accuracy of 99.05% was obtained with the mean absolute value (MAV) feature. The real-time classification was undertaken on eight subjects using the MAV feature's best performance with an average accuracy of 95.83% using the personalized data set and 91.79% using the generalized data set.ConclusionThe greatest accuracy is obtained by training the classifier with the subject's own data. Thus, it can be said that EMG signals are personal, just like fingerprints and retina. In addition, as a result, the tests repeated for 10 days showed the repeatability of the activation of the relevant muscle set and the training takes place and how this can be applied to those who will use prosthetic hands to obtain certain gestures.