藥物不良反造成了許多不必要的社會資源浪費，並且為病患及其家屬
帶來不必要的痛苦。目前已經有分許多人在嚐試想辨法供更加正確的相關資訊來
降低或避免藥物不良反應的發生。美國食品暨藥物管理(FDA)提供了一個藥物不
良反應的志願回報系統資料庫 (Spontaneous Reporting System Database
about ADR; AERS)，AERS 這個資料庫中包含了非常多有關藥物不良反應的 臨
床報告。在本論文中我們使用此FDA 所提供的資料庫並將重點放在找出哪些藥物
會由交互作用而產生不良反應。透過統計假設分析、交互作用其本身的特性跟決
策樹，我們可以找出藥物跟症狀之間的關聯性跟其它除了藥物本身之外的影響因
素，並對藥物交互作用產生更加精確的資訊。
統計假設分析可以用來表示藥物與症狀之間的關聯性，而交互作用
本身有的特性能幫我們評估一個症狀是由藥物交互作用所導致的可能性有多
少。最後透過決策樹，我們可以將其它因素考慮進來看哪些因素對不良反應有
影響，並幫助預測判別一個未知案例是否有可能發生不良藥物反應。

Adverse Drug Reaction (ADR) costs a lot of unnecessary social
recourse and leads to extra pain on patients. To provide the actual information
about ADR and avoid the rate of occurrence of ADR, many efforts have been
done. The US Food and Drug Administration (FDA) provide a Spontaneous
Reporting System Database about ADRs (AERS) which contains a lot of
clinical reports from about ADRs. In this study we focus on drug-drug
interaction caused ADRs. By using statistical hypothesis, characteristic of
interaction-caused ADRs, and decision tree, we can find the associations
between a set of drugs and symptoms, and related factors, than generate
more precisely signals of interactive drug pairs and symptoms related to them.
Statistical hypotheses testing can represent the association
between a set of drugs and a symptom, and the characteristic of drug-drug
interaction caused cases can help us evaluate the possibility of interaction. By
decision tree, we can take non-drug factors into consider and help the
prediction of unknown case.

[1] World Health Organization. “WHO Technical Report” 1972; 498

[2] Lazarou J, Pomeranz BH, Corey PN. “Incidence of adverse drug reactions in hospitalized patients: a meta-analysis of prospective studies”. JAMA. 1998; 279: 1200-5.

[3] Ana Szarfman, Stella G, Machado and Robert T. O’Neill. “Use of Screening Algorithms and Computer Systems to efficiently signal Higher-Than-Expected Combinations of Drugs and Events in the US FDA’s Spontaneous Reports Database”. Drug Safety, 2002; 25 (6) : 381-392

[4] Eugène P. van Puijenbroek, Antoine C.G. Egberts, Eibert R. Heerdink, Hubert G.M. Leufkens. “Detecting drug-drug interactions using a database for spontaneous adverse drug reactions: an example with diuretics and non-steroidal anti-inflammatory drugs”. European Journal of Clinical Pharmacology, 2000; 56 (9-10): 733-738, Springer Berlin / Heidelberg

[5] Eugène P. van Puijenbroek, Andrew Bate, Hubert G. M. Leufkens, Marie Lindquist, Roland Orre and Antoine C.G. Egberts. “A comparison of disproportionality for signal detection in spontaneous reporting systems for adverse drug reaction”. PHARMACOEPIDEMIOLOGY AND DRUG SAFETY, 2006; 11: 3-11

[6] William DuMouchel. “Bayesian Data Mining in Large Frequency Tables, with an Application to the FDA Spontaneous Reporting System”. The American Statistician, Vol. 53, No. 3. (Aug., 1999), pp. 177-190

[7] Lansner A, Hoist A. “A higher order bayesian neural network with spiking units”. Int J Neural Syst ,1996, 7:115-128

[8] A. Beta, M. Lindquist, I. R. Edwards, S. Olsson, R. Orre, A. Lansner, R. M. De Freitas. “A Bayesian neural network method for adverse drug reaction signal generation”. Eur J Clin Pharmacol (1998) 54: 315-321

[9] R.Orre, A. Lansner, A. Beta, M. Lindquist. “Bayesian nueral networks with confidence estimations applied to Data mining”. Computational Statistics & Data Analysis 34 (2000) 473-493

[10] WikiPedia
http://en.wikipedia.org/

[10] 行政院衛生署新竹醫院
http://dss.hch.gov.tw/other8.asp

[11] Christopher D. Manning, Hinrich Schütze. “Foundations of statistical natural language processing”. The MIT Press (1999)

[12] A. De Roeck, A. Sarkar and P. Garthwaite. "Defeating the Homogeneity Assumption", Proceedings of the 7th International Conference on the Statistical Analyisis of Textual Data (JADT 2004), Louvain La Neuve

[13] Adam Kilgarriff, Tony Rose. “Measures for corpus similarity and homogeneity”. 3 rd conference on Empirical Methods in Natural Language Processing, Granada, Spain, pp. 46 - 52

[14] drug digest
http://www.drugdigest.org/DD/Interaction/ChooseDrugs/1,4109,,00

[15] Simple Facts Sheet
http://www.atdn.org/sf.html