A Novel Margin Based Algorithm for Feature Extraction

Margin based feature extraction has become a hot topic in machine learning and pattern recognition. In this paper, we present a novel feature extraction method called Adaptive Margin Maximization (AMM) in which margin is defined to measure the discrimination ability of the features. The motivation comes principally from the iterative weight modification mechanism of the powerful boosting algorithms. In our AMM, the samples are dynamically weighted and features are learned sequentially. After one new feature is learned by maximizing the weighted total margin of data, the weights are updated so that the samples with smaller margins receive larger weights. The feature learned in the next round will thus try to concentrate more on these “hard” samples adaptively. We show that when the data are projected onto the feature space learned by AMM, most examples have large margins, and therefore the nearest neighbor classifier yields small generalization error. This is in contrast to existing margin maximization based feature extraction approaches, in which the goal is to maximize the total margin. Extensive experimental results on benchmark datasets demonstrate the effectiveness of our method.     

Adaptive binary tree for fast SVM multiclass classification

This paper presents an adaptive binary tree (ABT) to reduce the test computational complexity of multiclass support vector machine (SVM). It achieves a fast classification by: (1) reducing the number of binary SVMs for one classification by using separating planes of some binary SVMs to discriminate other binary problems; (2) selecting the binary SVMs with the fewest average number of support vectors (SVs). The average number of SVs is proposed to denote the computational complexity to exclude one class. Compared with five well-known methods, experiments on many benchmark data sets demonstrate our method can speed up the test phase while remain the high accuracy of SVMs.     

A novel plastic package for pressure sensors fabricated using the lithographic dam-ring approach

This study presents a novel plastic package for piezoresistive pressure sensors. A photoresist dam-ring patterned using the lithographic process is spin-coated on a piezoresistive pressure sensor to define a sensing channel in the pressure sensor package. Fluid epoxy molding encapsulates the pressure sensor and exposes the sensing channel during a high-temperature molding process at 165 °C. Experimental observations reveal that the silicon membrane of the pressure sensor is completely free of epoxy molding compound (EMC) contamination after the transfer molding process. The effectiveness of the dam-ring in shielding the silicon membrane of the pressure sensor during the molding process was confirmed. The packaged pressure sensor exerts a thermo-mechanical stress on the silicon membrane of the pressure sensor, resulting in an undesired output voltage drift. However, employing a package design with a large sensing channel opening can reduce the effect of package-induced stress. The proposed packaging scheme was a small package volume and surface-mount device (SMD) compatible features, making it suitable for portable commercial devices.     

Design and fabrication of novel micro electromagnetic actuator

This study designs, fabricates, and characterizes a novel micro electromagnetic actuator comprising a PDMS diaphragm, a polyimide-coated copper micro coil, and a permanent magnet. When an electrical current is passed through the micro coil, a magnetic force is induced between the coil and the magnet which causes the diaphragm to deflect, thereby creating an actuation effect. The experimental results demonstrate that the diaphragm deflection can be accurately controlled by regulating the current passed through the micro coil. It is shown that the maximum diaphragm deflection within elastic limits is 150 μm; obtained by passing a current of 0.6 A through a micro coil with a line width of 100 μm. The micro actuator proposed in this study is easily fabricated and is readily integrated with existing bio-medical chips due to its planar structure.     

Volatility model based on multi-stock index for TAIEX forecasting

Conventional time series models have been applied to handle many forecasting problems, such as financial, economic and weather forecasting. In stock markets, correct stock predictions will bring a huge profit for stock investors. However, conventional time series models produce forecasts based on some strict statistical assumptions about data distributions, and, therefore, they are not very proper to forecast financial datasets. This paper proposes a new forecasting model using adaptive learning techniques to predict TAIEX (Taiwan Stock Exchange Capitalization Weighted Stock index) with multi-stock indexes (NASDAQ stock index and Dow Jones stock index). In verification, this paper employs seven year period of TAIEX stock index, from 1997 to 2003, as experimental datasets, and the root mean square error (RMSE) as evaluation criterion. The performance comparison results show that the proposed model outperforms the listing methods in forecasting Taiwan stock market. Besides, from statistical test results, it is showed that the volatility of Dow Jones and the NASDAQ affect TAIEX significantly.     

Fault-tree analysis for liquefied natural gas terminal emergency shutdown system

Natural gas, one of the cleanest, most efficient and useful of all energy sources, is a vital component of the world’s supply of energy. To make natural gas more convenient for storage and transportation, it is refined and condensed into a liquid called liquefied natural gas (LNG). In a LNG site, safety is a long-team and critical issue. The emergency shutdown (ESD) system in the LNG receiving terminal is used to automatically stop the pumps and isolate the leakage section. Fault-tree analysis (FTA) has been widely used for providing logical functional relationships among subsystems and components of a system and identifying the root causes of the undesired failures in a system. In the conventional FTA for the ESD system, we usually assume that exact failure probabilities of events are collected. However, in most real applications, first, the FTA for the ESD system needs to be made at a early design or manufacturing stage, certain new components normally used without failure data; secondly, sometimes the environmental change in the system during the operation periods. This makes more difficult to gather past exact failures data for the FTA. To complete the FTA of the ESD system under these uncertain situations, we apply the intuitionistic fuzzy sets (IFS) theory to the FTA. We generate the intuitionistic fuzzy fault-tree interval, and the intuitionistic fuzzy reliability interval for the ESD system. We also present an algorithm to find the critical components in the system based on IFS–FTA and determine weak paths in the ESD system, where the key improvement must be made.     

基于面向方面MDA的实时系统开发方法

实时系统有许多难以处理的横切整个系统的非功能需求。针对以上问题,提出基于面向方面模型驱动架构的实时系统开发方法。该方法在设计阶段将系统的非功能需求从功能需求中分离出,以降低系统开发的复杂性,提高系统的可重用性、可维护性和横切关注点的模块化程度。通过应用实例证明该方法是有效的。     

基于可疑行为识别的PE病毒检测方法

针对当前PE病毒难以防范及查杀的现象,对PE病毒关键技术进行分析,提取病毒典型特征的可疑行为,在此基础上提出一种Windows平台下的静态检测方法。该方法在对程序反编译处理的基础上,以指令序列与控制流图的分析为行为识别依据,完成基于可疑行为识别的病毒检测方法的设计。实验结果证明,该检测方法能有效检测混淆变换病毒。     

基于多兴趣Agent的信息检索模型的研究

基于agent的信息检索系统在海量网络信息检索中已经得到广泛的应用.提出了一个基于多兴趣agent新的信息检索模型,该模型对查询信息进行了兴趣预测并生成了用户兴趣项权重向量.实验表明,模型在预测用户兴趣的效果和推荐搜索信息的准确率方面比传统的单兴趣Agent检索系统IRHOIA有一定的改善.