Feature selection and parameter optimization for support vector machines: A new approach based on genetic algorithm with feature chromosomes

Support vector machines (SVM) are an emerging data classification technique with many diverse applications. The feature subset selection, along with the parameter setting in the SVM training procedure significantly influences the classification accuracy. In this paper, the asymptotic behaviors of support vector machines are fused with genetic algorithm (GA) and the feature chromosomes are generated, which thereby directs the search of genetic algorithm to the straight line of optimal generalization error in the superparameter space. On this basis, a new approach based on genetic algorithm with feature chromosomes, termed GA with feature chromosomes, is proposed to simultaneously optimize the feature subset and the parameters for SVM.To evaluate the proposed approach, the experiment adopts several real world datasets from the UCI database and from the Benchmark database. Compared with the GA without feature chromosomes, the grid search, and other approaches, the proposed approach not only has higher classification accuracy and smaller feature subsets, but also has fewer processing time.     

Service improvement by business process management using customer complaints in financial service industry

In financial service industry, service improvement should be considered from process viewpoint and customer viewpoint because the value creation is ultimately linked with internal business processes on the back office and customers are involved as a co-producer of value. In this perspective, customer complaints through call centers are adequate to support the analysis for service improvement in financial service industry. In this study, we propose a web-based decision support system for business process management employing customer complaints, namely Voice of the Customer (VOC), and its handling data for service improvement. It involves VOC conversion for data enrichment and includes analysis of summarization, exception and comparison. The proposed system is evaluated on a major credit card company in South Korea.     

基于遥感的长沙市城市热岛效应时空分析

利用TM/ETM+数据,以长沙市为例,反演了地表温度,并针对不同时相的遥感数据,利用城市热岛强度来反映热岛效应强弱的变化。结果表明,长沙市的热岛空间分布与城市建成区的轮廓相吻合。城市热岛的范围随着城市建设、新建开发区以及道路交通网的发展不断增大,且在东南方向为主要增长方向。根据热岛强度鉴别并提取了强热岛区与正常区,发现长沙市强热岛区的面积已经从1993年的13.18km2扩大至2008年的68.16km2。最后,对长沙市热岛未来的发展趋势进行了预测。     

Low power integrated fluxgate sensor with a spiral magnetic core

Facing the new integrated applications, coordination between three parameters of the micro fluxgate sensor, namely, sensitivity, power consumption and operation range, is the key to make the applications come true. This paper reports a new low power micro integrated fluxgate sensor with low cost solution. A spiral electroplating permalloy magnetic core is introduced to reduce excitation current. Due to the design of little coils resistance and thick magnetic core, the fluxgate sensor exhibits a power consumption of 7.35 mW, a sensitivity of 117 V/T and a linear range of −200 to 200 μT. Compared with the constant-width magnetic core and the multi rectangular ring magnetic cores, experiment results show that the spiral magnetic core benefits both smaller excitation current and higher sensitivity. The newly developed fluxgate sensor maintains high sensitivity and wide linear range with low power consumption.     

Design and control of JAIST active robotic walker

This paper presents the design and control of a novel assistive robotic walker that we call “JAIST active robotic walker (JARoW)”. JARoW is developed to provide potential users with sufficient ambulatory capability in an efficient, cost-effective way. Specifically, our focus is placed on how to allow easier maneuverability by creating a natural interface between the user and JARoW. For the purpose, we develop a rotating infrared sensor to detect the user’s lower limb movement. The implementation details of the JARoW control algorithms based on the sensor measurements are explained, and the effectiveness of the proposed algorithms is verified through experiments. Our results confirmed that JARoW can autonomously adjust its motion direction and velocity according to the user’s walking behavior without requiring any additional user effort.     

Observer-based networked control for continuous-time systems with random sensor delays

This paper is concerned with the networked control system design for continuous-time systems with random measurement, where the measurement channel is assumed to subject to random sensor delay. A design scheme for the observer-based output feedback controller is proposed to render the closed-loop networked system exponentially mean-square stable with H_∞ performance requirement. The technique employed is based on appropriate delay systems approach combined with a matrix variable decoupling technique. The design method is fulfilled through solving linear matrix inequalities. A numerical example is used to verify the effectiveness and the merits of the present results.     

Editorial for the SI: Optimization of Discrete Event Dynamic Systems

We consider it a great privilege to introduce this special two-issue series on Optimization of Discrete Event Dynamic Systems. Optimization has played a central role in the research discipline of discrete event dynamic systems since its inception, and has been the focus of many papers published in this journal since its creation. Driven by applications in several areas such as manufacturing, transportation, and telecommunications, the optimization problem has been addressed by our research community with particular emphases on theory, computation, and simulation. The purpose of this special issue is to bring together papers representing diverse problems of current interest, and to expose and highlight state-of-the-art ideas, techniques, and algorithms that have been developed for their solution.     

基于监控器时间开销的虚拟机发现方法

针对传统方法只能发现单一类型虚拟机的缺陷,提出基于虚拟机监控器时间开销的虚拟机发现方法。特定指令能使监控器运行时产生显著的额外开销,该方法能利用监控器执行不同指令序列产生的相对时间开销对虚拟机进行判别。实验结果表明,该方法能够准确发现目前3类主流虚拟机。     

CAN节点总线关闭状态下通信中断的解决方法

CAN总线通信中,当某一节点进入总线关闭状态后,不会与总线其他节点通信.针对这个问题,分析了节点从总线关闭状态转化为主动错误状态的条件,提出了采用模拟开关和电压基准主动实现这种转化条件的解决方法,解决了CAN节点恢复时间不确定的问题.     

Estimation of winter wheat phenology under the influence of cumulative temperature and soil salinity in the Yellow River Delta,China, using MODIS time-series data

Understanding the impact of environmental factors on crop phenology is significant in predicting crop growth stages, agricultural decision-making, and yield estimation. Here, using Moderate Resolution Imaging Spectroradiometer time-series data, we present phenological detection mechanisms and an explanation for the phenological variability linked to environmental drivers, such as cumulative temperature and soil salinity, for winter wheat (Triticum aestivum L.) in the Yellow River Delta in 2013. The 8-day normalized difference vegetation index was fitted to a double Gaussian function. Phenological phases, such as the green-up and heading phases, were extracted using maximum curvature approaches. The spatial characteristics of the phenological patterns were investigated. The relationships between the phenological phases and cumulative temperature were explored. Then, the relationships between the phenological phases and soil salinity were evaluated by selecting sites with similar soil fertility and temperature forcing. This study concluded that the regional average green-up date occurred on 5 March, and the regional average heading date occurred on 9 May. The spatial distributions of the green-up and heading phases showed a gradual delay from the southwest to the northeast and from the south to the north. The green-up phase lagged 4–5 days for every 10 degree days that the cumulative temperature decreased. The heading phase lagged 1–2 days for every 10 degree days that the cumulative temperature decreased. The green-up phase in a non-salinization region might be approximately 5–9 days earlier than that in a severe or moderate salinization region. The heading phase in a severe region might occur approximately 1–8 days earlier than that in a non-salinization or moderate salinization region. The method proposed in this article may be useful for understanding the impact of temperature and soil salinity on phenology and could be used to better manage winter wheat in coastal salinization areas.