The rapid development of technology and economy has largely enhanced the quality of life. Nevertheless, various social and environmental problems have emerged. It would be the key solution to develop environmental education in order to have people present the environmental knowledge and the attitudes and value to concern about the environment and develop the skills and action to solve environmental problems. Environmental education teachers in 25 colleges and universities in central and southern Taiwan are researched in this study. Total 250 copies of questionnaires are distributed, and 224 valid copies are retrieved. The research results conclude the significant positive correlations 1.between professional development and professional knowledge & competence, 2.between teacher belief and professional knowledge & competence, and 3.between teaching efficacy and professional knowledge & competence. According to the research result, suggestions are eventually proposed in this study. It expects to arouse the public cognition of environment to care for the environment through curriculum instruction of environmental education teachers and have people present the environmental knowledge and the attitudes and value to concern about environment.

Malicious script,such as JavaScript, is one of the primary threats of the network security. JavaScript is not only a browser scripting language that allows developers to create sophisticated client-side interfaces for web applications, but also used to carry out attacks taht used to steal users' credentials and lure users into providing sensitive information to unauthorized parties. We propose a static malicious JavaScript detection techniques based on SVM(Support Vector Machine). Our approach combines static detection with machine learning technique, to analyze and extract malicious script features,and use the machine learning technology,SVM, to classify the scripts.This technique has the characteristics of high detection rate,low false positive rate and the detection of unknown attacks. Applied to experiments on the prepared data set, we achieved excellent detection performance.