올해 여름, KAIST에서 계산사회과학(Computational Social Science, CSS)에 관심 있는 분들을 위한 집중 프로그램인 <2026 SICSS-Korea (Summer Institute in Computational Social Science)>를 KDI School과 함께 개최할 예정입니다(with 윤지성 교수님, 조성아 교수님).
최근 들어 계산사회과학은 그 어느 때보다 중요해지고 있습니다. 이는 새로운 방법론의 등장 때문만이 아니라, 사람, 정보, 기술, 제도 등이 서로 얽혀 만들어내는 사회적 과정을 사회과학의 이론적 관점과 데이터·컴퓨팅을 결합하여 이해하는 것의 중요성이 매우 커지고 있기 때문입니다. 특히나 인공지능이 이러한 사회적 과정 위에서 작동하고 이를 다시 재구성하는 핵심 기술로 자리 잡으면서, 이러한 이해의 필요성은 아주 커지고 있습니다.
이런 맥락에서 계산사회과학은 사회과학 연구에 컴퓨팅을 적용하는 것을 포함하는 건은 물론이거니와, 사람, 정보, 기술, 제도 등이 어떻게 상호작용하는지를 이해하려는 다학제적 시도입니다. 정치학, 지리학, 사회학, 커뮤니케이션, 경제학 등 다양한 사회과학 분야와, 산업공학, 컴퓨터과학, 정보학, 네트워크과학 등의 통찰을 연결하며, 그동안 분절적으로 다뤄져 왔던 연구 흐름들을 하나의 관점으로 통합합니다.
—
이번 프로그램에는 국내외에서 활동하는 계산사회과학 분야의 탑 클래스 연구자들이 참여하여 강연을 진행할 예정입니다. 최신 연구 흐름과 방법론을 직접 접할 수 있는 기회가 될 것입니다. 이와 더불어, 이번 프로그램은 단순한 강의 중심 프로그램에 그치지 않고, 참가자들이 직접 아이디어를 발전시키고 함께 연구를 논의하는 작고 밀도 있는 1주 프로그램으로 구성되어 있습니다.
약 15명 내외로 소수 인원을 선발하며, 선발된 참가자에게는 KAIST 기숙사 숙박 및 다수의 식사가 지원됩니다(단, 이동 비용은 지원되지 않습니다).
다음은 주요 일정입니다. 특히 4월 23일 2시에 온라인 설명회가 예정되어 있으니 많이 참여 부탁 드립니다(RSVP 링크 및 줌 링크는 포스터 참고):
Social Network Analysis and Text Mining for Big Data: The Power of Words and Networks提出了前沿的方法和工具，弥合了文本挖掘与社交网络分析研究之间的鸿沟，同时也为分析大规模文本和网络数据提供了新的见解。
《面向大数据的社会网络分析与文本挖掘：语言与网络的力量》（Social Network Analysis and Text Mining for Big Data: The Power of Words and Networks）提出了前沿的方法和工具，弥合了文本挖掘与社交网络分析研究之间的鸿沟，同时也为分析大规模文本和网络数据提供了新的见解。这些工具旨在满足商业分析师和研究人员的需求，以助力开创性分析结果的产生。
Andrea Fronzetti Colladon在罗马第三大学（Roma Tre University）担任教授，并负责领导商业与集体智能实验室（Business and Collective Intelligence Lab）。他致力于研究组织与个人的沟通、决策及变革模式，通过解码复杂系统为管理决策提供科学依据。他的研究融合了网络科学、自然语言处理和机器学习方法，并与社会科学、心理学和人文学科理论相结合，体现了深厚的跨学科底蕴。他发明了语义品牌评分（Semantic Brand Score）和OCEAN™变革管理模型（OCEAN™ Change Management Model），旨在将学术洞察转化为组织领导者可用的实践工具。其学术足迹涵盖了与麻省理工学院集体智能中心（MIT Center for Collective Intelligence）、东北大学（Northeastern University）和西奈山医院（Mount Sinai Hospital）等国际顶尖机构的广泛合作。作为多部学术著作的作者，他的代表作品包括本书以及《引领有意义的变革》（Leading Meaningful Change），这些著作为理解语言与网络在数据分析中的力量提供了重要框架。此外，他还通过教学与咨询服务于TIM（TIM）、Enel（Enel）和萨奇广告（Saatchi & Saatchi）等知名企业，助力团队应用创新管理工具并在复杂环境中提升竞争力。
罗伯托·韦斯特雷利
Roberto Vestrelli
Roberto Vestrelli现为意大利佩鲁贾大学（University of Perugia）工业与信息工程专业的博士候选人。他与安德烈亚·弗龙泽蒂·科拉东教授（Andrea Fronzetti Colladon）合著了《大数据的社会网络分析与文本挖掘：词汇与网络的力量》（Social Network Analysis and Text Mining for Big Data: The Power of Words and Networks）一书，展示了其在大数据分析领域的学术造诣。罗伯托·韦斯特雷利（Roberto Vestrelli）的研究方向集中于社会网络分析、文本挖掘及其在商业、环境治理与社会科学中的应用。他的研究成果涵盖了气候风险信息披露对企业价值的影响、媒体在塑造能源叙事中的作用，以及大语言模型在评估新闻自由时的表现等前沿课题。此外，他还参与开发了针对GDELT数据库（GDELT）的文本重构工具，并致力于通过大数据方法分析非洲地区的外国直接投资决定因素。通过将语义品牌评分等量化指标应用于数字营销与服务管理，他为理解复杂网络中的言论力量提供了重要的技术支持与理论视角。
二
出版信息
书名：《面向大数据的社会网络分析与文本挖掘：语言与网络的力量》（Social Network Analysis and Text Mining for Big Data: The Power of Words and Networks）
语言和网络在塑造历史、文化和人类行为方面拥有巨大的力量。本引言章节探讨了词汇如何通过精心构建的叙事产生激发情感反应、驱动社会变革或引发冲突的潜力，正如 Franco-Prussian War 期间的 Ems Dispatch 所证明的那样。与此同时，通过 Social Network Analysis (SNA) 描绘的人类联系揭示了维系社会、影响领导力并推动运动的隐藏结构。当 Text Mining 与 SNA 的技术工具相结合时，会产生极具价值的洞见，从而使人们能够更深入地理解决策过程和社会动态。基于这一观点，本章介绍了支持复杂系统研究的分析工具，阐明了语言和网络如何塑造人类互动以及社会结果。
Words and networks hold immense power in shaping history, culture, and human behavior. This opening chapter explores how words, through carefully crafted narratives, have the potential to evoke emotional responses, drive societal change, or initiate conflicts, as demonstrated by the Ems Dispatch during the Franco-Prussian War. At the same time, (human) connections – mapped through social network analysis (SNA) – reveal the hidden structures that bind societies, influence leadership, and drive movements. Invaluable insights emerge when the techniques and tools of text mining and SNA are combined, enabling a deeper understanding of decision-making processes and social dynamics. In line with this view, this chapter introduces analytical tools to support the study of complex systems, shedding light on how words and networks shape human interaction and societal outcomes.
This chapter introduces fundamental and advanced (social) network analysis concepts and builds a foundation for understanding how networks help study complex phenomena. It begins with defining networks and their main components, such as nodes and edges. The chapter explains how to represent and analyze social phenomena using data from interviews or unstructured sources. It introduces centrality metrics, enabling readers to identify strategically positioned nodes within a network, and examines clusters and communities, focusing on techniques to detect them in a graph. The discussion extends to identifying key roles within a social system and understanding the implications of positional changes in a network. The chapter also explores dynamic network analysis, outlining methods for examining time-evolving interaction dynamics and introducing metrics such as rotating leadership. Building on the authors’ expertise, this chapter offers a practical, research-driven overview of the most impactful metrics and techniques used in their scientific work. Rather than aiming for exhaustive coverage, it focuses on equipping readers with essential tools needed to grasp and apply both foundational and advanced concepts.
This chapter offers a comprehensive introduction to text mining and natural language processing (NLP), emphasizing practical techniques for analyzing large-scale unstructured text data. It covers core methods like data preprocessing, sentiment analysis, and named entity recognition, with a particular focus on constructing and interpreting word networks. We present basic and advanced approaches for evaluating textual features, such as novelty, informativeness, and complexity. We also highlight the innovative application of social network analysis to examine word relationships, offering a unique perspective on textual data analysis.
This chapter explores the transformation of textual data into networks and the benefits of analyzing them using social network analysis techniques. The chapter emphasizes the power of combining text mining and network analysis to understand relationships between words and how this approach can be applied to fields such as marketing and natural language processing. The chapter also introduces the Semantic Brand Score, a composite indicator designed to measure the importance and memorability of brands or concepts within textual data by leveraging network analysis. Lastly, we demonstrate practical applications of these methods, including brand image analysis and a novel approach to topic modeling.
第五章
文本挖掘与社会网络分析的集成：案例研究与应用
本章通过四个案例研究阐明了整合网络分析与文本挖掘技术的价值。第一个案例研究利用 New York Times 的在线新闻文章，评估了四个科技品牌的 Semantic Brand Score。第二个案例研究探讨了人工智能对劳动力动态的影响，其数据基础源于专利和职位描述。第三个案例研究聚焦于从 Glassdoor 收集的某全球咨询公司员工评论，展示了识别组织优劣势感知的有效方法。最后一个案例研究则通过挖掘 New York City 的 Airbnb 住客评论，利用非结构化文本勾勒出该市的各类城市特征。虽然目前已有大量用于 Social Network Analysis 和 Text Mining 的软件包，以及 R 和 Python 等编程语言的函数库，但能有效融合这些方法的图形界面应用仍较为匮乏。由本书作者之一开发的 SBS BI 软件集成了前述章节讨论的网络分析、文本挖掘及词网络分析技术，为研究者提供了针对这些方法的集成化解决方案。
This chapter presents four case studies that illustrate the value of integrating network analysis and text mining techniques. The first case study assesses the Semantic Brand Score of four technology brands by analyzing online news articles from the New York Times. The second case study explores the influence of artificial intelligence on workforce dynamics, using textual data from patents and job descriptions. The third case study focuses on employee reviews for a global consulting firm collected from Glassdoor, showcasing methods to effectively identify perceived organizational strengths and weaknesses. The final case study examines Airbnb guest reviews in New York City, leveraging unstructured text to map various urban characteristics of the city. While numerous tools and software packages are available for social network analysis and text mining, along with extensive libraries for programming languages like R and Python, few graphical interface applications effectively combine these methodologies. The SBS BI software, developed by one of the book’s authors, incorporates the network analysis, text mining, and word network analysis techniques discussed in the previous chapters, providing an integrated approach to these methods.
语义学研究的是语言中的意义，研究词语和短语如何表达思想、传达观念。但正如前文所指出的，人类语言不仅具有语义，也具有结构，也就是词语彼此如何关联、如何连接。词语会受到其所处语境、句法和语法结构的塑造。例如，请看这个句子：The bank closed at 5 PM。bank 这个词可以指金融机构、河岸，甚至可以指飞机的一个倾斜动作。语义学帮助我们识别这些可能的意义，但句子的结构及其中其他词语，尤其是动词 closed，则提示这里所说的是一家金融机构。类似 at 5 PM 这样的时间表达，又进一步澄清了这一解释。由此可见，一个词的意义，是通过它与其他词语的互动而形成的。因此，文本分析不能排除对词语连接关系的研究，因为词语的意义往往与其关系网络不可分割。
将社会网络分析与文本挖掘整合起来，为理解人类行为带来了一个新的维度。网络并不仅仅关乎谁与谁发生连接，正如词语也绝不仅仅是孤立的发声。二者结合起来，构成了一幅由互动、影响与意义交织而成的复杂织锦。通过同时分析个体、组织，甚至观念之间的关系，以及它们所使用的语言，我们就能够以前所未有的方式洞察信息如何流动、决策如何形成，以及人们如何回应其周围世界。例如，通过结合网络与词语的力量，我们就能够描绘、分析并理解媒体如何塑造某一位领导人或某一政治党派的形象。当某个政治人物在语义网络中处于中心位置时，他可能会在公民心中留下强烈印象。设想一个政治品牌频繁出现在新闻报道中，并经常与深嵌于媒体话语中的多样化主题或概念联系在一起，例如慈善、可持续性或移民议题。这个政治品牌就可能在潜在选民心中脱颖而出。实际上，只要观察媒体话语所生成的词语网络，便有可能预测选举结果。然而，网络分析与语义分析的结合，并不限于政治领域。这种方法可以应用于广泛的行业与场景。以推动能源转型这一挑战为例：我们如何确保像 energy community 或 green transition 这样的概念能够在消费者心中引发共鸣，并进而影响其行为？如果仅仅依赖这些术语出现的频率，一场单纯的媒体宣传攻势可能远远不够。真正使这些概念具有影响力的，是意义与词语连接方式的结合——也就是，这些信息如何与共同价值、日常关切或未来愿景发生联系。某些词语被如何、有意识地安放在媒体话语结构之中，往往决定了一场行为变革是否能够成功；正如精确校准过的修辞既可能引发冲突，也可能促进和平。
Winnie E. Pérez Martínez’s post on the Scholars’ Lab blog this week got me thinking. In “Breadth and Depth, a Self-Centered Dialectic,” she revisits how we discuss breadth and depth as two approaches to digital humanities professional development. In this framing, one that I have put forward myself, we can think of careers in DH as operating on two axes. On the one, we are expected to know a little about a lot of things. On the other, we are directed more towards narrow, specialist-level knowledge about a smaller subset of methods. Breadth vs. depth. Few careers really ask us to go entirely in both directions. More practically, we tend to specialize in a couple areas within DH and develop passing familiarity with many more.
For me, the dichotomy between breadth and depth was a way to help students map their career plans onto the different skills they might acquire. I thought of it as a way to free yourself from the need to be expert in everything. In her post, Pérez Martínez expertly shows how breadth and depth actually inform and lead to one another. There can be no one right way in. If you start deep, you might find yourself broadening, and starting wide can help you to focus in. What most resonated about Pérez Martínez’s post, though, was the way in which you can see an exceptional scholar and practitioner wrestling over whether they are enough, over whether they could ever develop the necessary skills they need to feel complete. Those anxieties never really go away. I feel them too. I recognized myself in Pérez Martínez’s post, and I couldn’t help but sense that the breadth against depth framing seemed to be having the opposite effect I would want, heightening anxiety rather than mitigating it.
Pérez Martínez proposes a broadening of the axes I had envisioned. Breadth and depth move beyond just X and Y, curling in upon themselves until they start to push outwards. The moment reminded me of the age-old dichotomy of “hack” vs. “yack” in DH work and how Laura Braunstein offered “stack” as an important third term. In addition to coding and technological critique as key parts of DH work, Braunstein’s intervention elevates “the often invisible technological, social, and physical structures within which scholarship is produced and disseminated.” For Braunstein, DH work is more than just the sum of what we do, it also consists of the structures we put in place to enable that work. In the same spirit and inspired by Pérez Martínez, I have been wondering what breadth and depth leave out, what they gesture towards within and beyond the teaching that we do.
Put another way, what is education if not just content? One point of comparison here is L. Dee Fink, whose Taxonomy of Significant Learning illuminates the various components of teaching.
Fink’s Taxonomy usefully illustrates all the things that lie beyond the subject matter in the courses we teach. Learning is more than consuming books, articles, or topics. Teaching is more than passing along skills and methods. If we think of DH merely as skill building, we live too much in the upper right of the circle. We leave out the rest of what makes DH experiences—and DH learning—significant for so many of us. We ignore the transformative mentoring that shows a variety of career options. We miss the collaborative practices that can change how we view our work in dialogue with others. We do not account for how true interdisciplinarity changes our perspectives on our own research processes. We need a new term to trouble the dichotomy between breadth and depth that can capture a more capacious view of what it means to practice digital research and teaching, one that goes beyond subject matter, methods, and skills.
I find this particularly urgent in the age of generative AI, a complicated set of technologies that threatens to instrumentalize education beyond recognition. What counts as methodological training if you can vibe code your way to a launched digital project? What counts as digital pedagogy if our students are secretly using chatbots as study partners? How do we make room for conversations about professional development that do not reduce people to a tidy axis of skill acquisitions?
What lies beyond the breadth and depth of what it means to be a digital humanist?
I would introduce a third term for DH professional development: “breath.” Breadth and depth ask us to think about what we can and cannot do, about the subject matter and methods of DH work. The terms ask us to think about the limits of our knowledge and our inability to pursue universal expertise. Breath asks us to reframe the conversation entirely. It is an invitation to pause and re-embed our work in the body. How do we feel about our labor? Who are the working souls in DH and how do we engage with them? How do we work or overwork our own body to the point of breathlessness? What is the lived experience of our labor that transcends the skills or methods? What are the affects—the joys, frustrations, traumas, triumphs—of DH work that cannot be captured by thinking in terms of skill acquisition? How do our energies map onto a living, breathing community of thinkers and doers beyond the work on the table in front of us? Where do we fit in?
Breadth and depth ask students to think about where they could be, professional development by way of spatial orientation. Breath invites students to consider where they are, to think of themselves as real people with real needs that need attending.
The Digital Humanities Alliance for Research and Teaching Innovations (DHARTI) is seeking proposals from educational institutions and/or stakeholders from the G.L.A.M. [galleries, libraries, archives, and museums ]sector in India to host the DHARTI 2026 biennial conference scheduled to take place in December 2026. Based on the previous conferences organised under the aegis of DHARTI, DHARTI […]
The Roberto Busa Prize is an award of the Alliance of Digital Humanities Organisations (ADHO). It is named in honour of Father Roberto Busa, the first pioneer of humanities computing, who in 1949 began experiments in linguistic automation, with the support of the IBM offices in New York and Milan, as part of his analytical… Read More »Call for Nominations: ADHO Roberto Busa Prize 2026
Lately, I’ve been thinking a lot about that canonical debate around depth vs. breadth in digital humanities (DH). I remember initially my reaction was feeling that “true” DHers should be both, that this was an aptitude anyone could and should develop in the spirit of the discipline.1 After much method trial-and-error, I then grew frustrated that I hadn’t found “my thing” and disappointed to think I could only do DH in breadth, not depth. I was bound to a path instead of choosing one. Eventually, I decided I didn’t care for the division at all.
Lately, though, I’ve started wondering, what if the question has multiple axes acting simultaneously? What if breadth leads to depth, the opposite happens, or something else altogether? What if, instead of choosing a path, it emerges, for each one of us, from our own individual way of sharing, problem-solving, organizing, and researching?
In my “failing” to become an expert at one single thing calling out to me, I had no choice but accepting and recognizing the value of breadth, and inevitability of failure—another core DH moment—even if my perceived failure was personal in nature.2 But the beauty of “rock bottoms” is that they create an opportunity for radical change because one’s ego is finally humble enough to give up on what isn’t working and engage in earnest reflection, even start over. ON/OFF. Unplug and replug. It’s a time to accept losses and reassess them against growth, assets, skills, experience, effort, connections. What did you learn in the process? In DH, the process is key.3
Through the process, I learned I’ve been wrong all along: depth & breath are really two sides of the same coin. What separates us should be instead the base principles for a constructive bridge uniting DH inquiry. I used to think that not knowing and being new to the tools of an environment was a disadvantage (breadth+1), but it turns out specialists (depth+1) often don’t remember what a beginner feels or needs, precisely because they have intimate knowledge of their fields. The specialist (depth+1) is always iterating over their craft, perfecting techniques or creating new ones, deeply connected to the innovations of peers, organizations, and entities influencing their work. Generalists (breadth+1), typically, pay attention to what’s going on in multiple fields, who’s working where, what is the budget distribution, how do you get more students in to apply for your programs, how do you process their applications. DH emerges from a delicate act of balancing out critical responsibilities among experts of both kinds, each deeply attuned to their own and each other’s tasks.
Revisiting this discussion with recent insights, I realized that, as far as the depth & breadth goes, I am through and through a digital humanist in breadth (of course, an oxymoron). This is how my brain works. I have a good instinct for quickly finding unusual connections between topics, and a curiosity for learning anything, so I end up knowing a little bit about many things, not much about anything at all. I used to badly resent this indecisiveness in my learning pattern, the inability to choose and stick to one thing passionately forever, but I have been reworking this perspective to understand that what I am is flexible, curious, nimble. This is a quality that makes me good at handling unknown situations, at learning how to learn new things on my own, and at translating that knowledge of the learning process to expert audiences.
And yet, over time, I have found myself going back to specific projects, questions, topics. I keep thinking about design, minimal computing, pedagogy, infrastructures, technology, small datasets, and collaborative work, for example, over and over again in recent years. These are now inescapable research areas for me. Won’t this path of ceaseless returning eventually lead to depth as projects grow and change from recurring personal concerns, interests? Maybe we all become breadth/depth hybrids if we spend long enough earnestly engaging DH work, which is kind of pretty, it means connection awaits at the center where both axes intersect.
Figure 1. A simple diagram of breadth and depth as perpendicular intersecting axes.
In schnellen Schritten geht es auf die DHd2027 in Marburg zu. Daher möchten wir uns an dieser Stelle noch einmal herzlich für die Konferenzbegleitungsbeiträge unserer Reisestipendiat:innen der DHd2026 bedanken. Stipendien gab es für Doktorand:innen und (very) early career DHers. Sie wurden vom DHd-Verband sowie von NFDI4Culture, NFDI4Memory und CLARIAH-AT vergeben.
Die Stipendiat:innen konnten im Gegenzug die Konferenz entweder auf Social Media begleiten oder haben im Anschluss an die Tagung jeweils einen Blogpost verfasst, der die breite Palette an Themen und Ansätzen auf der DHd dokumentiert. Nachfolgend findet sich eine Übersicht über die verschiedenen Blogposts mit den dazugehörigen Links:
Karin Wallner, Wie steht es um die Sichtbarkeit von Frauen und Queerness in den Digital Humanities? Ein Blick in die DHd-Jahreskonferenz 2026 in Wien, 06. März 2026, in: DHd Blog, https://dhd-blog.org/?p=23392
Pierre-Michel Weiße, Abstract, Review, Postersession – (M)ein Weg zur DHd2026, 27. März 2026, in: DHd Blog, https://dhd-blog.org/?p=23544
Diese Blogposts in ihrer Themenvielfalt zeigen die Bandbreite der Methoden und Ansätze in den Digital Humanities im deutschsprachigen Raum. Die vielseitigen Beiträge der Stipendiat:innen sind ein wertvoller Beitrag zur Dokumentation der Diskurse und Erlebnisse der DHd2026, wofür wir uns ganz herzlich bedanken!
Das vierte Community Meeting 2026 von NFDI4Objects findet virtuell statt! Die Veranstaltung richtet sich an Einsteiger:innen und Interessierte und bietet spannende Einblicke in die Arbeit des Konsortiums. Freut euch auf interaktive Formate, Austauschmöglichkeiten und dezentrale Online-Workshops.
Programm und Workshop-Termine folgen in Kürze – merken Sie sich den Termin schon jetzt vor!
Hiermit laden wir Euch sehr herzlich zum nächsten Clustertreffen des CC „Authority Files and Community-driven Vocabularies“ am 29. April 2026, von 13 bis 14:30 Uhr, ein. Dieses Mal wird es ein Hands on für DANTE mit unserem Chair Michael Markert von der VZG geben.
DANTE ist ein Webservice zur Pflege und Veröffentlichung von Vokabularen, der von der Verbundzentrale des GBV auch für NFDI4Objects-Institutionen bereitgestellt wird. Zu unserem Hands-On-Treffen wollen wir gemeinsam Datensätze im DANTE-Testpool anlegen und bearbeiten, damit neue Nutzer:innen sich mit der Funktionalität vertraut machen können.
Bringt gern Beispieldaten aus dem Bereich Personen, Körperschaften, Orte und Sachbegriffe zum Ausprobieren mit!
Es wäre schön, wenn auch erfahrene DANTE-Nutzer:innen zur Unterstützung bei Fragen dabei sein könnten.
Am Dienstag, 28. April 2026 von 9 bis 11 Uhr findet die nächste Sitzung des Community Clusters Protected Heritage Sites online statt.
Themen der Sitzung sind:
Abschluss: Datenmodelle für Maßnahmenobjekte und Schutzflächen
Voraussichtlich wird dies die letzte Sitzung sein, in der die Modelle Thema sind. Falls Sie keine Zeit haben, an dem Treffen teilzunehmen, schicken Sie daher gerne Ihre Anmerkungen per E-Mail, damit sie in der Diskussion berücksichtigt werden können.
Das Protokoll der 15. Sitzung des CC Protected Heritage Sites vom 12.03. ist ab sofort via OSF verfügbar: https://osf.io/pwzg4/. Hier finden Sie auch die aktuellen Entwürfe der Datenmodelle.
登录
俄 - 西伯利亚联邦大学（DHlab Siberian）
英 - 谢菲尔德大学（Sheffield）
5 - ZfdG (Zeitschrift für digitale Geisteswissenschaften)
4 - 德语区数字人文研究协会博客（DHd-Blog）
