運用Facebook公開資料監測類流感疫情

Detecting influenza epidemics using Facebook public data

柳姚仁(Yao-Jen Liu)

指導教授 ： 張昭憲

淡江大學/商管學院/資訊管理學系碩士在職專班/碩士(2015年)

https://doi.org/10.6846/TKU.2015.00988

摘要

流行性感冒每年都會帶來數十萬人的死亡，造成嚴重的健康威脅。流感可透過接種疫苗降低疫情，若能提早發現流感散播的趨勢，便可降低感染與死亡人數。臺灣目前無較為即時的流感監測系統，疾管署的類流感監測系統根據急診資料統計，會有一週至一個月的延滯時間，對於流感爆發早期預警而言，時效性明顯不足。
　　本研究希望能透過網路社群留言分析，發展一套具有高即時的流感監測方法。首先，我們透過特定關鍵字組合，分析Facebook上與流感相關之公開訊息，對比疾管局的流感統計數據，透過相關性分析，建立以社群網路為基礎之流感疫情監測模型。實驗結果顯示，此權重模型預測之數據與官方統計數字有顯著之相關，證實使用社群網站監測流感疫情的可行性。根據本論文之研究成果，希望能做為政府監測流感疫情的先期指標，降低流感風險。

關鍵字

皮爾森相關係數 ； 相關性分析 ； 流感監測 ； 社群網路分析

並列摘要

Every year, influenza causes hundreds of thousands deaths, resulting in serious health threat. Flu epidemic can be reduced through vaccination, if early detection of influenza spread trends, infection and deaths can be reduced. Taiwan currently no immediate flu surveillance system to monitor influenza-like illness system according to the emergency department statistics, there will be a week to a month's delay time, for early warning of a pandemic, the timeliness is clearly insufficient.
This study try to analyze the message from the social media to develop a set of high instantaneous influenza surveillance methods. Through a specific keyword combinations to analysis the public data related with flu on Facebook , compared CDC's influenza statistical data through correlation analysis, we establishment an influenza surveillance model that based on social network.
Experimental results show that this model predicts the weight data of the official statistics have significant correlation, confirming the feasibility of using social network to monitor flu. According to results of this research, the hope we can do for the government to monitor early indicators of flu activity, reduce flu risk.

並列關鍵字

Pearson's correlation coefficient ； correlation analysis ； influenza surveillance ； social network analysis

參考文獻

[8] 衛生福利部疾病管制署，傳染病介紹＞流感，網址http://www.cdc.gov.tw/diseaseinfo.aspx?treeid=8d54c504e820735b&nowtreeid=dec84a2f0c6fac5b&tid=3013B7FC8F965336 ，上網日期：2015/02/24

[9] Akay, A., et al., “A Novel Data-Mining Approach Leveraging Social Media to Monitor Comsumer Opinion of Sitagliptin, “ IEEE Journal of Biomedical and Health Informatics, Vol. 19, No. 1, Jan. 2015, pp. 389-396.

[15] Farley, Monica M. MD, “2009 H1N1 Influenza: A Twenty-First Century Pandemic With Roots in the Early Twentieth Century”, American Journal of the Medical Sciences, Volume 340 - Issue 3, 2010, Page 202-208

[16] Fox, S., “Online Health Search 2006” , Pew Internet & American Life Project (2006)

[17] Hulth, A., Rydevik, G. & Linde, A., “Web Queries as a Source for Syndromic Surveillance”, PLoS ONE 4(2): e4378. doi:10.1371/journal.pone.0004378 (2009).