武漢肺炎接觸史小工具專案

馬上用 Google Maps 了解過去14天內有沒有和 COVID-19 擦肩而過

網站 Website

參與開發 Github

說明文件 README

專案簡介 Project Info

緣由 Background

詳細作法 Detail Design

預定使用者 Target Users

現有類似專案 Prior Works

相關專案 Related Works

授權方式 License

資料來源 Data Source

專案目前狀態 Current State

Data

i18n

利益揭露 Disclosure of Interests

徵求協作者 Collaboration

聯絡方式 Contact Us

Todo List

成果展示 Website

專案簡介 Project Info

緣由 Background

武漢肺炎正在流行,各國衛生單位在難以掌握案例接觸史的情況下,可能會公布該案例的移動軌跡。此專案希望提供一個線上工具讓民眾可以簡單的比對自己的歷史軌跡是否有和案例的軌跡重疊。

國際交流如此頻繁的今天,單獨一個國家的防疫並不足夠,這個系統希望能成為各國政府以及各國民眾都可以使用的平台。

There are more and more COVID-19 cases in several countries. CDC of each country might release the trace of a confirmed case if CDC couldn’t identify the contact history. This project provides an online tool for users to compare their own trace with the traces of confirmed cases.

Nowadays, people can easily travel internationally. All countries have to work together to prevent the outbreak of diseases. We hope this project can become a system that helps everyone and every country in the world.

詳細作法 Detail Design

Google Maps 使用者如果開啟 Location History 功能,就可以在 Google Maps Timeline 頁面看到自己的歷史軌跡。使用者可以將自己的歷史軌跡和衛生單位公布的案例歷史軌跡互相比對,判斷自己和該案例接觸的可能性有多高。但這個過程十分的麻煩,如果案例數增加,使用者可能會放棄比對,或是用非常寬鬆的方式估計自己的風險,得到錯誤的結果(偽陽性或偽陰性都有可能)。

使用者可以從 Google Maps Timeline 下載自己某日的歷史軌跡(KML格式),可以使用程式幫忙比對使用者軌跡和案例軌跡是否有重疊。用程式來判斷的好處是,我們可以控制比較基準,判斷的結果會更一致。

專案以網站作為使用者介面,

  1. 使用者造訪網站時,網站會下載案例的軌跡到本地端。
  2. 網站會引導使用者到 Google Timeline 下載自己的歷史軌跡(儲存於本地端)。
  3. 使用者藉由 file selector 選擇想要比對的軌跡,瀏覽器會讀取這些軌跡,並和案例軌跡比對。此過程完全在本地端完成,使用者的軌跡不會被上傳到任何地方
  4. 網站顯示比對結果,列出可能的接觸地點和時間。

Google 另外提供了一個 Takeout 服務打包自己的資料,其中一項也有歷史軌跡檔(中文叫「定位記錄」,是 JSON 格式)。

If Location History is enabled on Google Map, users can find their location history on Google Timeline. Users can compare their history with the trace released by authorities, and they can estimate their likelihood of contacting a confirmed case. However, this is a tedious process, if there are a lot of cases, span through multiple days and multiple locations, the users might give up and/or determine their likelihood with some (random) heuristics. The result would be flaky, the rate of false positive and false negative will be high.

Users can download their own history from Google Timeline (in KML format). Technically, we can parse their KML file and compare the trace with the traces of confirmed cases. The benefit of doing the comparison by a program is, the quality of results would be consistent. We can control / estimate the rate of false positives and false negatives.

The project will be presented by a website.

  1. When a user visits the website, the website downloads traces of confirmed cases and save locally.
  2. The webpage guides the user to Google Timeline to download their location history of given days. The downloaded file is saved locally and is only saved locally.
  3. The user uses a file selector on the website to select a list of their traces to compare. The webpage loads and parses the traces, and compares them with the traces of confirmed cases. During the process, user data is not uploaded to any server.
  4. Finally, the webpage shows potential matches on the page, each match contains a time period and location.

Google also provides the Takeout service. The service allows you to download your location history, in JSON format.

預定使用者 Target Users

想要比對自己移動軌跡和確診病例移動軌跡的人。

People who want to compare their traces and traces of confirmed cases.

現有類似專案 Prior Works

目前沒有,歡迎幫忙補充

Not as we know of.

相關專案 Related Works

授權方式 License

資料來源 Data Source

專案目前狀態 Current State

已上線(見成果展示

Released, please check the Website section.

Data

資料部分有台灣和韓國部分案例資料,請見資料來源

Please check the Data Source section.

i18n

目前有中文、英文、韓文版本。語言選擇只會影響頁面呈現的效果,所有資料在所有語言中都可以使用。

利益揭露 Disclosure of Interests

(牽涉到哪些組織團體、有哪些已知的或潛在的金錢或物質或無形利益報酬)

就我們所知沒有......

Not as we know of...

徵求協作者 Collaboration

聯絡方式 Contact Us

Todo List

成果展示 Website