基于线性回归和属性集成的分类算法

对于高维度小样本数据的分类问题,高维属性的复杂性限制了分类模型预测的准确率。为了进一步提高准确率,提出了基于线性回归和属性集成的分类算法。首先,采用线性回归为每一个属性构建属性线性分类器(Attribute Linear Classifier,ALC);其次,为了避免因ALC数量过多而导致准确率下降,利用经验风险最小化策略中的经验损失值作为评估标准来优选ALC;最后,应用多数投票法来集成被筛选的ALC。采用高维度小样本的基因表达数据集进行实验,结果显示该算法具有比逻辑回归、支持向量机和随机森林算法更高的准确率。

在多数据库集成环境中的属性集成方法的研究

描述了一种基于模糊集合理论的属性集成方法，它可显式地表示数据库集成系统中数据属性值的不一致程度。介绍了语义更为丰富的扩充的结构化查询语言。最后结合例子描述了在多数据库集成环境中一种属性集成方法。

基于属性集成的三支聚类算法

在三支聚类中,每一个类簇由核心域、边界域、琐碎域三个集合来表示,体现出聚类中的不确定性,降低了决策风险。针对属性数较多的数据集,本文提出了一种基于属性集成的三支聚类算法。该算法通过对样本随机选择部分属性使用kmeans算法进行聚类,并根据每次聚类的结果采用集成策略将元素划分至核心域、边界域或琐碎域中。在UCI等数据集上测试的结果表明,该算法相较于传统的kmeans算法,可以较好地提升聚类结果的ACC、DBI和AS指标。

基于量子云模型反馈的协同精英属性均衡优势集成约简

为进一步增强进化算法在最小属性约简过程中的全局求解性能,提出了一种基于量子云模型反馈的协同精英属性均衡优势集成约简算法,该算法首先设计一种基于云模型反馈的量子自适应旋转角调整策略,使量子蛙群精英在云模型定性知识和罚因子反馈指导下自适应控制属性搜索空间范围;然后构建一种有限理性区域下协同精英均衡优势属性分解框架,在动态精英演化区域内使参与属性约简的量子蛙群精英在平均权重裕度下协同化达到Nash均衡优势区域;最后量子蛙群精英采用集成化操作机制在各自均衡优势区域内协同提取属性约简子集,从而稳定取得全局最优约简特征集。实验结果表明本算法求解全局最优属性约简集效率、精度和稳定性等具有明显优势,应用到孕龄新生儿脑MRIs电子病历分割,进一步表明该算法具有较强的应用性能。

多属性决策的组合赋权优化方法

针对多属性决策中属性权重的确定问题,提出了一种主观权重与客观权重的集成方法。对每个决策属性,将各个决策者给出的主观权重和客观权重作为样本点,采用规划的方法来集成主观权重和客观权重,使决策结果更加可靠、可信。

商业组件的集成可信度模型研究

除了增强功能,组件开发者还应该提高组件的质量;组件用户则根据组件的集成可信度,评估组件库中的可选组件,从而选择合适的组件,减少系统集成过程中的风险.从组件的可靠性、可用性、可测试性、安全性、维护性五类属性来评估,并详细考察了组件的一些子属性,采用概率的经验方法来建立组件的集成可信度模型,提供了一种可以直接数值比较的度量方法.

依据主、客观权重集成最终权重的一种方法

本文根据已掌握的主、客观信息,利用最优化技术构造既反映主观信息,又反映客观信息的确定属性权重的最优化模型,通过求解该优化模型得出各属性的最终权重．

Linguistic approaches to multiple attribute decision making in uncertain linguistic setting

We studied multiple attribute decision-making problems with uncertain linguistic information, in which the preference values took the form of uncertain linguistic variables. We introduced some operational laws of uncertain linguistic variables and a formula for the comparison between two uncertain linguistic variables. We proposed two new aggregation operators called extended uncertain linguistic aggregation (EULA) operator and interval linguistic aggregation (ILA) operator, and then develop an EULA operator-based linguistic approach and an ILA operator-based linguistic approach, respectively, to multiple attribute decision making in uncertain linguistic setting. The approaches were straightforward and do not produce any loss of information. Finally, an illustrative example was given to verify the developed approaches and to demonstrate their practicality and effectiveness.

Co-evolutionary cloud-based attribute ensemble multi-agent reduction algorithm

In order to improve the performance of the attribute reduction algorithm to deal with the noisy and uncertain large data, a novel co-evolutionary cloud-based attribute ensemble multi-agent reduction（CCAEMR） algorithm is proposed.First, a co-evolutionary cloud framework is designed under the M apReduce mechanism to divide the entire population into different co-evolutionary subpopulations with a self-adaptive scale. Meanwhile, these subpopulations will share their rewards to accelerate attribute reduction implementation.Secondly, a multi-agent ensemble strategy of co-evolutionary elitist optimization is constructed to ensure that subpopulations can exploit any correlation and interdependency between interacting attribute subsets with reinforcing noise tolerance.Hence, these agents are kept within the stable elitist region to achieve the optimal profit. The experimental results show that the proposed CCAEMR algorithm has better efficiency and feasibility to solve large-scale and uncertain dataset problems with complex noise.

Amino Acid Composition and Functional Properties of Giant Red Sea Cucumber (Parastichopus californicus) Collagen Hydrolysates

Giant red sea cucumber (Parastichopus californicus) is an under-utilized species due to its high tendency to autolysis.The aim of this study was to evaluate the functional properties of collagen hydrolysates from this species.The degree of hydrolysis (DH),amino acid composition,SDS-PAGE,emulsion activity index (EAI),emulsion stability index (ESI),foam expansion (FE),and foam stability (FS) of hydrolysates were investigated.The effects of pH on the EAI,ESI FE and FS of hydrolysates were also inves-tigated.The results indicated that the β and α1 chains of the collagen were effectively hydrolyzed by trypsin at 50℃ with an En-zyme/Substrate (E/S) ration of 1:20 (w:w).The DH of collagen was up to 17.3% after 3 h hydrolysis with trypsin.The hydrolysates had a molecular weight distribution of 1.1 17 kDa,and were abundant in glycine (Gly),proline (Pro),glutamic acid (Glu),alanine (Ala) and hydroxyproline (Hyp) residues.The hydrolysates were fractionated into three fractions (< 3 kDa,3 10 kDa,and > 10 kDa),and the fraction of 3 10 kDa exhibited a higher EAI value than the fraction of > 10 kDa (P<0.05).The fraction of > 10 kDa had higher FE and FS values than other fractions (P<0.05).The pH had an important effect on the EAI,ESI,FE and FS.All the fractions showed undesirable emulsion and forming properties at pH 4.0.Under pH 7.0 and pH 10.0,the 3 10 kDa fraction showed higher EAI value and the fraction of > 10 kDa showed higher FE value,respectively.They are hoped to be utilized as functional ingredients in food and nutraceutical industries.

Development of Integrated Precision Vibration-Assisted Micro-Engraving System

A novel precision vibration-assisted micro-engraving system was developed by the integration of fast tool servo and ultrasonic elliptical vibration system, in which the flexure hinge was designed to avoid backlash and PID control algorithm was established to guarantee specific precision. Apart from experimental validation of the performance of the system, various micro-V-grooves cutting experiments on aluminum alloy, ferrous material and hard cutting material were performed, in which Kistler force sensor was used to measure cutting force. Through experiments, it was clear that the vibration-assisted micro-engraving system can ensure good quality of micro-V-grooves and reduce cutting force by about 60% compared with traditional removal process without ultrasonic vibration.

自然保护区生态地理信息系统开发研究

为服务于珠峰国家级自然保护区的生态信息管理与评价,设计开发了一套综合性的生态地理信息系统。描述了系统的实施方案和技术特色,重点强调了有关生态的专题属性集成,并提出了基于GIS地理处理建模的生态分析、评价和预测的模型驱动方案,对自然保护区管理以及生态问题的深入研究提供了持续的信息化保障。

Facile synthesis of poly(aroxycarbonyltriazole)s with aggregation-induced emission characteristics by metal-free click polymerization

Regioseletive 1,3-dipolar polycycloadditions of 4,4'-isopropylidenediphenyl dipropiolate (1) and tetraphenylethene (TPE)-containing diazides (2) are carried out in polar solvents such as DMF/toluene at a moderate temperature of 100℃ for 6 h,producing poly(aroxycarbonyltriazole)s (PACTs) P3 with high molecular weights (Mw up to 23900) and regioregularities (F1,4 up to ～90%) in high yields (up to ～99%).These metal-free click polymerizations can propagate smoothly in an open atmosphere without protection from oxygen and moisture.The obtained polymers are soluble in common organic solvents and thermally stable at temperatures up to 375℃.Thanks to their contained TPE moieties,the PACTs show aggregation-induced emission and can serve as fluorescent chemosensors for superamplified detection of explosives.