管制圖分析應用於腦炎文獻資料

Abstract

現今科技進步迅速，讓資料的藏儲變的越來越容易，資料量自然也越來越龐大，如何從這些大量資料中萃取出用戶所需的資訊，便成為近年來熱烈討論的議題。在本論文中，我們以PubMed這個生醫資料庫中的腦炎資料為對象，利用管制圖做為分析工具，期望能藉此找到一些關於腦炎發生率的資訊。在此，我們主要研究三種腦炎，分別為：Japanese encephalitis、Herpes simplex encephalitis和 Limbic encephalitis，並將這些腦炎資料運用多種不同的管制圖進行分析。首先會以individual and moving Range(MR)charts對三種類型腦炎文獻的數量資料繪製管制圖，試圖找出會產生這些疾病的重要資訊，並運用其他統計方法如時間序列，James-Stein Estimator等方法來協助做出更正確的管制圖監控法，期望發現其他有益資訊。由研究結果發現，本論文提出的方法，對瞭解疾病資訊上有著一定的幫助，不過也需要相關專業知識及好的原始資料。

With the rapid development of technology, it is more convenient to store data, and consequently, the amount of data that is being created and stored keeps growing.A famous biomedical data source, The PubMed, is a free search engine accessing primarily the MEDLINE database of references and abstracts on life sciences and biomedical topics. In this study, we access the literatures of encephalitis from PubMed and adopt control chart methods to monitor the number of literatures with respect to time frame. We consider the three types of encephalitis, Japanese encephalitis, Herpes simplex encephalitis and Limbic encephalitis in this study. The individual and moving range (MR) control charts, as well as, Hotelling Tsquare control charts, time series analysis and the James-Stein estimator, are adopted to analyze the data. The results show that this proposed method may be a useful approach to analyze the occurrence rate of these diseases.