標題: | 疾病模型上的個體差異與行為:結核病電腦模擬 Personal diversity and behavior on epidemiology model:A computer simulation on Tuberculosis (TB) |
作者: | 林彥廷 Yen-Ting Lin 孫春在 Chuen-Tsai Sun 資訊科學與工程研究所 |
關鍵字: | 結核病模型;疾病模擬;具分身點概念細胞自動機;小世界網路;TB model;disease simulation;cellular automata with mirror identity model;small-world network |
公開日期: | 2005 |
摘要: | 本研究提出一個以代理人為基礎,並結合人際接觸社會網路的結核病電腦模擬模型。根據結核病的傳播特性,該模擬模型可分成三個層面:個體、個體行為、社會接觸。模型底層的人際接觸社會網路是一個小世界複雜網路,具有高群聚度與低分隔度二種在真實社會中常見的拓樸特性。我們利用代理人來模擬真實社會中的人類個體,並利用代理人本身的屬性與規則設定來表現人類個體間的差異(如年齡、性別、抵抗力等)與行為。透過模型的三個層次,我們整合家庭群聚概念、社會分身點概念、多代理人系統、細胞自動機來模擬結核病的二種主要的傳播模式─透過人際間的「親密」與「隨意」接觸所造成的感染。之後,我們使用真實的結核病流行曲線等資料來驗證模擬模型,透過此模型平台,瞭解結核病在傳染的各種巨觀與微觀現象,推估未來傳播趨勢與可能的結果,並嘗試探討公共衛生政策與醫療行為介入的模擬結果。 We propose a agent-based tuberculosis(TB) transmission model combine contact network. According the tuberculosis properties, We can separate our model into three levels (1)Individual (2)Individual behavior and(3)Social contact, The social network we used has small-world properties that is high degree of clustering and low degree of separation. We use agent-based system to realize personal diversity(ex: age,gender and immunity) and behavior. Through these three components we combine house-structure, cellular automata, multi-agents system with mirror identity model to reflect TB transmission through person’s close contact and casual contact. We use real data to validate our model. We can predict the trand of TB and evaluate the intervence of health policies and behavior. |
URI: | http://140.113.39.130/cdrfb3/record/nctu/#GT009323576 http://hdl.handle.net/11536/79104 |
顯示於類別: | 畢業論文 |