A scalable kernel-based semisupervised metric learning algorithm with out-of-sample generalization ability

In recent years, metric learning in the semisupervised setting has aroused a lot of research interest. One type of semisupervised metric learning utilizes supervisory information in the form of pairwise similarity or dissimilarity constraints. However, most methods proposed so far are either limited to linear metric learning or unable to scale well with the data set size. In this letter, we propose a nonlinear metric learning method based on the kernel approach. By applying low-rank approximation to the kernel matrix, our method can handle significantly larger data sets. Moreover, our low-rank approximation scheme can naturally lead to out-of-sample generalization. Experiments performed on both artificial and real-world data show very promising results.     

A new graph cut-based multiple active contour algorithm without initial contours and seed points

This paper presents a new graph cut-based multiple active contour algorithm to detect optimal boundaries and regions in images without initial contours and seed points. The task of multiple active contours is framed as a partitioning problem by assuming that image data are generated from a finite mixture model with unknown number of components. Then, the partitioning problem is solved within a divisive graph cut framework where multi-way minimum cuts for multiple contours are efficiently computed in a top-down way through a swap move of binary labels. A split move is integrated into the swap move within that framework to estimate the model parameters associated with regions without the use of initial contours and seed points. The number of regions is also estimated as a part of the algorithm. Experimental results of boundary and region detection of natural images are presented and analyzed with precision and recall measures to demonstrate the effectiveness of the proposed algorithm.     

Test frame updating in CPM testing of Prolog programs

Category Partition Method (CPM) is a general approach to specification-based program testing, where test frame reduction and refinement are two important issues. Test frame reduction is necessary since too many test frames may be produced, and test frame refinement is important since during CPM testing new information about test frame generation may be achieved and considered incrementally. Besides the information provided by testers or users, implementation related knowledge offers alternative information for reducing and refining CPM test frames. This paper explores the idea by proposing a call patterns semantics based test frame updating method for Prolog programs, in which a call patterns analysis is used to collect information about the way in which procedures are used in a program. The updated test frames will be represented as constraints. The effect of our test frame updating is two-fold. On one hand, it removes “uncared” data from the original set of test frames; on the other hand, it refines the test frames to which we should pay more attention. The first effect makes the input domain on which a procedure must be tested a subset of the procedure’s input domain, and the latter makes testers stand more chance to find out the faults that are more likely to show their presence in the use of the program under consideration. Our test frame updating method preserves the effectiveness of CPM testing with respect to the detection of faults we care. The test case generation from the updated set of test frames is also discussed. In order to show the applicability of our method an approximation call patterns semantics is proposed, and the test frame updating on the semantics is illustrated by an example.

基于ASP.NET的水稻分子育种信息平台管理系统

为实现水稻生物学数据的共享、交流和利用,基于ASP.NET,结合国内外水稻分子育种研究的现状和需求,构建了水稻分子育种信息平台.系统平台设计了基因库、QTL库和Ontology等多个子库和模块,并针对生物学数据的特点,采用B/S架构,运用存储过程和分页技术等多重手段,实施了客户端和服务器端双重安全屏障.对系统的实现技术、实现方案和优化设计等进行了详细的讨论和阐述.     

基于LM3S316的微控制器仿真USB设备

提出了基于LM3S316实现与PC机通讯的方法.给出了硬件实现及软件设计思想.该方法使用微控制器仿真USB设备,降低了功耗、成本.     

A novel fault-tolerant execution model by using of mobile agents

The exponential expansion of the Internet and the widespread popularity of the World Wide Web give a challenge to experts on reliable and secure system design, e.g., e-economy applications. New paradigms are on demand and mobile agent technology is one of the features. In this paper, we propose a fault-tolerance execution model by using of mobile agents, for the purpose of consistent and correct performance with a required function under stated conditions for a specified period of time. Failures are classified into two classes based on their intrinsic different effects on mobile agents. For each kind of failure, a specified handling method is adopted. The introduction of exceptional handling method allows performance improvements during mobile agents’ execution. The behaviors of mobile agents are statistically analyzed through several key parameters, including the migration time from node to node, the life expectancy of mobile agents, and the population distribution of mobile agents, to evaluate the performance of our model. The analytical results give new theoretical insights to the fault-tolerant execution of mobile agents and show that our model outperforms the existing fault-tolerant models. Our model provides an effective way to improve the reliability of computer systems.     

Ubiquitous Personal Study: a framework for supporting information access and sharing

The information resources on the Web are diversified, the amount of which is increasing rapidly. Demands for selecting useful information from the Internet, managing personal contents, and sharing contents under control have risen. In this study, we propose the Ubiquitous Personal Study, a framework of personalized virtual study to support accessing, managing, organizing, sharing and recommending information. In this paper, we focus on discussing the framework, and design and implementation issues on how to implement it with Web 2.0 mash-up technology and Open Source Software.     

压缩应力下（Tb,Dy）Fe2磁致伸缩的研究

本文利用自制的磁致伸缩测量仪和“ｊｕｍｐ”效应测量仪研究了在一定压强了多晶（Ｔｂ、Ｄｙ）Ｆｅ２棒的磁致伸缩,发现加压时,在一定工艺条件下制作的（Ｔｂ、Ｄｙ）Ｆｅ２棒会有磁致伸缩的变化。卸载后经过热处理,磁致伸缩值在不加压时也会维持一个高值。     

过敏原肌质钙结合蛋白在毕赤酵母中的表达、优化及免疫反应性研究

目的 探究天然肌质钙结合蛋白(sarcoplasmic calcium binding protein, SCP)的可替代物,为蟹类过敏原的检测提供基础材料,本研究首次利用毕赤酵母(Pichia pastoris, P. pastoris)高效表达表达三疣梭子蟹(Portunus trituberculatus)重要过敏原SCP,并检验其免疫反应性。方法 根据毕赤酵母的密码子偏好性优化SCP基因并构建重组质粒。将其热激转化至P. pastoris GS115菌株后经遗传霉素(Geneticin, G418)筛选获得阳性高拷贝子。最后通过甲醇诱导表达重组SCP并结合免疫印记(Western blotting, WB)和间接酶联免疫吸附实验(enzyme-linked immunosorbent assay, ELISA)验证其免疫反应性。结果 SCP在P. pastoris GS115中实现了可溶性高效表达,其表观分子量约为28 kDa。在摇瓶水平下,最佳诱导条件为pH为6.0、每24 h添加1.0%(v/v)甲醇,于28℃发酵144 h,在此条件下,纯度为91.6%的SCP产量可达15 mg/L。WB和间接ELISA结果表明,重组SCP具有IgG结合能力。结论 毕赤酵母表达系统可以得到纯度较高且免疫反应性良好的重组SCP。本研究为SCP的理化研究及产业化应用奠定了基础,并有望促进特异性甲壳类过敏原检测的发展。