Handbook of Computational Social Science for Policy

This open access handbook describes foundational issues, methodological approaches and examples on how to analyse and model data using Computational Social Science (CSS) for policy support. Up to now, CSS studies have mostly developed on a small, proof-of concept, scale that prevented from unleashing its potential to provide systematic impact to the policy cycle, as well as from improving the understanding of societal problems to the definition, assessment, evaluation, and monitoring of policies. The aim of this handbook is to fill this gap by exploring ways to analyse and model data for policy support, and to advocate the adoption of CSS solutions for policy by raising awareness of existing implementations of CSS in policy-relevant fields.

To this end, the book explores applications of computational methods and approaches like big data, machine learning, statistical learning, sentiment analysis, text mining, systems modelling, and network analysis to different problems in the social sciences. The book is structured into three Parts: the first chapters on foundational issues open with an exposition and description of key policymaking areas where CSS can provide insights and information. In detail, the chapters cover public policy, governance, data justice and other ethical issues. Part two consists of chapters on methodological aspects dealing with issues such as the modelling of complexity, natural language processing, validity and lack of data, and innovation in official statistics. Finally, Part three describes the application of computational methods, challenges and opportunities in various social science areas, including economics, sociology, demography, migration, climate change, epidemiology, geography, and disaster management.

The target audience of the book spans from the scientific community engaged in CSS research to policymakers interested in evidence-informed policy interventions, but also includes private companies holding data that can be used to study social sciences and are interested in achieving a policy impact.

這本開放存取手冊描述了計算社會科學（CSS）在政策支持方面的基礎問題、方法論和分析建模數據的示例。迄今為止，CSS研究主要在小規模的概念驗證上進行，這限制了其發揮系統性影響力以支持政策循環，並改善對社會問題的理解，從而對政策的定義、評估、評價和監測產生影響。本手冊的目的是填補這一空白，通過探索分析和建模數據以支持政策的方法，提倡在政策領域中采用CSS解決方案，以提高對CSS在政策相關領域中現有實施的認識。

為此，本書探討了計算方法和方法，如大數據、機器學習、統計學習、情感分析、文本挖掘、系統建模和網絡分析，對社會科學中不同問題的應用。本書分為三個部分：第一部分的基礎問題章節首先介紹和描述了CSS可以提供洞察和信息的關鍵政策制定領域。具體而言，這些章節涵蓋了公共政策、治理、數據正義和其他倫理問題。第二部分包括處理方法論方面的章節，涉及複雜性建模、自然語言處理、數據的有效性和缺乏、以及官方統計數據的創新等問題。最後，第三部分描述了計算方法在各種社會科學領域中的應用、挑戰和機遇，包括經濟學、社會學、人口學、移民、氣候變化、流行病學、地理學和災害管理。

本書的目標讀者群包括從事CSS研究的科學界社群，對以證據為基礎的政策干預感興趣的政策制定者，以及持有可用於研究社會科學並有興趣實現政策影響的私營公司。

Eleonora Bertoni is a Project Officer - Computational Social Scientist at the European Commission, Joint Research Centre (JRC), where she works for the Centre of Advanced Studies (CAS). She coordinates the activities of the CAS group on Computational Social Science for Policy which aims at building capacity in accessing and analysing non-traditional data, as well as exploring applications of computational methods in different social sciences domains to address specific policy questions.

Matteo Fontana is a Project Officer - Data Scientist at the Joint Research Centre of the European Commission. His main research interest is the application and development of data science and statistical learning techniques to evaluate complex data sources in the social sciences field. He is particularly interested in nonparametric inference and prediction, with a focus on conformal methods for complex data. From an applicative point of view, he is interested in macro-economic forecasting, migration modelling and environmental economics.

Lorenzo Gabrielli is a Data Scientist in the JRC Centre for Advanced Studies (CAS) Project on Computational Social Science for Policy to carry out scientific tasks, i.e. harness non-traditional data including big data, analyse it and draw conclusions on its impact on society. He has gained experience in the analysis of big data with data mining and machine learning techniques in national and international contexts by collaborating with several public and private research institutes.

Serena Signorelli works as a Data Partnership and Management Officer at the Joint Research Centre. Her current project focuses on Computational Social Science for Policy, and it is part of the Centre for Advanced Studies of the Scientific Development unit. Her research interests have mainly focused on the use of Wikipedia page views to study tourism flows, and they have been exploited through a traineeship and a subsequent contract with the Eurostat Big Data task force.

Michele Vespe is a Team Leader at the European Commission, Joint Research Centre, where he coordinates the activities of teams of researchers for investigating societal consequences associated with the improved availability of digital trace data, including research in the fields of data governance. He also leads the Computational Social Science for Policy project team.

Eleonora Bertoni是歐洲委員會聯合研究中心（JRC）的計算社會科學項目主任，她在高級研究中心（CAS）工作。她負責協調CAS計算社會科學組的活動，該組旨在建立能力以訪問和分析非傳統數據，並探索在不同社會科學領域應用計算方法來解決特定政策問題。

Matteo Fontana是歐洲委員會聯合研究中心的數據科學家項目主任。他主要研究興趣是應用和發展數據科學和統計學習技術，以評估社會科學領域中的複雜數據來源。他對非參數推斷和預測特別感興趣，並專注於複雜數據的符合方法。從應用的角度來看，他對宏觀經濟預測、移民建模和環境經濟學感興趣。

Lorenzo Gabrielli是JRC高級研究中心（CAS）計算社會科學項目的數據科學家，負責進行科學任務，即利用非傳統數據（包括大數據）進行分析，並對其對社會的影響做出結論。他通過與多個公共和私營研究機構的合作，在國內外背景下獲得了使用數據挖掘和機器學習技術分析大數據的經驗。

Serena Signorelli在歐洲委員會聯合研究中心擔任數據合作和管理主任。她目前的項目專注於計算社會科學為政策，並且是科學發展單位高級研究中心的一部分。她的研究興趣主要集中在使用維基百科頁面訪問量研究旅遊流量，並通過實習和隨後與歐盟統計局大數據任務組的合同加以利用。

Michele Vespe是歐洲委員會聯合研究中心的團隊負責人，他負責協調研究團隊的活動，研究與數字跟踪數據的改善可用性相關的社會影響，包括數據治理領域的研究。他還領導計算社會科學為政策項目團隊。