- 期刊
毒品尿液篩檢程序所產生問題與試劑所影響之閾值標準值誤差實證研究：以貝氏定理公式計算
Using Bayesian Probability to Increase the Accuracy of Urine Testing for Drugs
摘要
於篩檢毒品當中尿液為目前國內主要檢驗毒品方法之一，惟其準確度卻常因各種篩劑內容或各種人為、自然因素受到極大挑戰，故無法以單純篩檢陽性結果立即得證是否有無使用毒品。基於此現象，本研究試著以數學公式中之貝氏定理（Bayes' theorem）之條件機率（Bayesian probability）下進行演算如何於初篩程式當中正確演算出毒品篩檢之結果，同時以文獻探討與量化分析等實證研究方法達成以下研究目的並導出結論：其一、探討執行尿液篩檢程序所產生之問題與試劑所影響之閾值標準值誤差；其二、貝氏定理中之條件機率如何影響篩檢準確度之分析；其三、以量化問卷分析找出影響毒品篩檢之準確度因素。
並列摘要
The abuse of narcotics and pharmaceutical drugs is a major public security problem in all advanced countries, and urine testing is currently the main method for detecting drug abuse. However, the accuracy of urine screening is affected by a variety of factors, both natural and contrived. Based on these considerations, in this study we use such empirical research methods as document analysis, quantitative analysis, and Bayesian probability to determine ways to improve the accuracy of drug screening as well as the effectiveness of various measures for preventing drug abuse. The purposes of this research are as follows: 1) to clarify the procedures used for urine testing and their level of accuracy; 2) to use Bayesian probability to determine how various factors affect screening accuracy; and 3) to use quantitative analysis to determine ways for increasing the accuracy of urine screening.