chore: add citations May 2025

Author: sebastian-nagel

bib/cc2017.bib

@@ -221,8 +221,25 @@ @InProceedings{cc:MehmoodShafiqWaheed:2017:regional-context-www
   year         = "2017",
   booktitle    = "2017 IEEE 13th Malaysia International Conference on Communications (MICC)",
   title        = "Understanding Regional Context of World Wide Web using Common Crawl Corpus",
+  pages        = "164--169",
+  doi          = "10.1109/MICC.2017.8311752",
   URL          = "https://www.researchgate.net/publication/321489200_Understanding_Regional_Context_of_World_Wide_Web_using_Common_Crawl_Corpus",
   pdf          = "https://www.researchgate.net/profile/Amir_Mehmood/publication/321489200_Understanding_Regional_Context_of_World_Wide_Web_using_Common_Crawl_Corpus/links/5a251abaaca2727dd87e780a/Understanding-Regional-Context-of-World-Wide-Web-using-Common-Crawl-Corpus.pdf",
+  abstract     = "The World Wide Web has emerged as the most important and essential tool for the society. Today, people
+                 heavily rely on rich resources available in the web for communication, business, maps, and social
+                 networking etc. In addition, people seek web content in their preferred regional language besides
+                 English. The global statistics of the world wide web are well known, however, the regional context of
+                 the world wide web is poorly understood. This paper presents large scale web study using Common Crawl
+                 Corpus of December 2016. We examine 200+ terabytes of data with Amazon's Elastic MapReduce
+                 infrastructure. We analyze 2.87 billion web documents with respect to content type, domains, and
+                 content language. Furthermore, we explore multi-lingual web pages for European and Asian languages. Our
+                 results show that 97.8\% of web documents present in our data are “text/html”. In addition, 57.2\%
+                 of web documents contain content in the English language. Moreover, web content in Russian language has
+                 5.7\% share which is more that any other European language. Furthermore, we found that 60.6\% of web
+                 documents have content exclusively in the English language. Finally, we found that Japanese and
+                 traditional Chinese language content dominate the Asian web pages with 1.89\% and 1.23\% share. To the
+                 best of our knowledge, this is the first large scale web study to explore the language mix present in
+                 the web documents.",
   cc-dataset-used = "CC-MAIN-2016-50",
   cc-statistics = "languages, multi-lingual content, MIME types, TLDs, web server",
   cc-processing-tools = "EMR (AWS grant), CLD2",

bib/cc2022.bib

@@ -388,21 +388,21 @@ @Article{cc:KreutzerCaswellWangWahabEtAl:2022:audit-web-multilingual-datasets
                  Alessia and Baruwa, Ahmed and Bapna, Ankur and Baljekar, Pallavi and Azime, Israel Abebe and Awokoya,
                  Ayodele and Ataman, Duygu and Ahia, Orevaoghene and Ahia, Oghenefego and Agrawal, Sweta and Adeyemi,
                  Mofetoluwa",
-  title        = "{Quality at a Glance: An Audit of Web-Crawled Multilingual Datasets}",
+  title        = "Quality at a Glance: An Audit of Web-Crawled Multilingual Datasets",
   journal      = "Transactions of the Association for Computational Linguistics",
   volume       = "10",
   pages        = "50--72",
   year         = "2022",
   month        = "01",
-  abstract     = "{With the success of large-scale pre-training and multilingual modeling in Natural Language Processing
+  abstract     = "With the success of large-scale pre-training and multilingual modeling in Natural Language Processing
                  (NLP), recent years have seen a proliferation of large, Web-mined text datasets covering hundreds of
                  languages. We manually audit the quality of 205 language-specific corpora released with five major
                  public datasets (CCAligned, ParaCrawl, WikiMatrix, OSCAR, mC4). Lower-resource corpora have systematic
-                 issues: At least 15 corpora have no usable text, and a significant fraction contains less than 50\\%
+                 issues: At least 15 corpora have no usable text, and a significant fraction contains less than 50\%
                  sentences of acceptable quality. In addition, many are mislabeled or use nonstandard/ambiguous language
                  codes. We demonstrate that these issues are easy to detect even for non-proficient speakers, and
                  supplement the human audit with automatic analyses. Finally, we recommend techniques to evaluate and
-                 improve multilingual corpora and discuss potential risks that come with low-quality data releases.}",
+                 improve multilingual corpora and discuss potential risks that come with low-quality data releases.",
   ISSN         = "2307-387X",
   doi          = "10.1162/tacl_a_00447",
   URL          = "https://doi.org/10.1162/tacl\_a\_00447",

bib/cc2023.bib

@@ -2002,8 +2002,7 @@ @Misc{cc:Murray:2023:Generative-AI-Art-Copyright
                  one of the images generated by the AI as the final image. The end-users then make further decisions
                  about the actual use and its function and purpose for the images the end-users selected and adopted
                  from the outputs of the AI. In the course of working with the AI tool to try to produce a certain
-                 image, an end-user might steer the system to produce a work that could,
-?? BibTeX string too long for field ``abstract''.
+                 image, an end-user might steer the system to produce a work that could, 
  under an infringement analysis,
 		  be regarded as potentially infringing, which would lead us
 		  again to the fair use analysis based on the end-user’s
@@ -2031,3 +2030,46 @@ @Misc{cc:Murray:2023:Generative-AI-Art-Copyright
   cc-author-affiliation={University of Kentucky, USA},
   cc-class	= {legal/copyright, legal/fair-use, nlp/language-model, ai/foundation-model, nlp/multi-modal-language-model, nlp/multimodal-corpora, ai/image-dataset}
 }
+
+
+@Article{cc:Kriesch:2023:Web-Mining-und-NLP-in-der-Wirtschaftsgeographie,
+  title        = "Web {Mining} und {Natural} {Language} {Processing} als methodisches {Komplement} in der
+                 {Wirtschaftsgeographie}",
+  URL          = "https://jlupub.ub.uni-giessen.de//handle/jlupub/16306",
+  abstract     = "Für wirtschaftsgeographische Forschung spielen räumlich und inhaltlich granular aufgelöste Daten
+                 eine zentrale Rolle, um Treiber und Barrieren sozioökonomischer Entwicklungen von Regionen besser
+                 verstehen zu können. Vor dem Hintergrund der zunehmenden Digitalisierung hat sich das Internet zu
+                 einer enorm umfassenden Datenquelle für unterschiedlichste For-schungsdisziplinen entwickelt.
+                 Insbesondere die Fähigkeit moderner Algorithmik auch un-strukturierte Textdaten semantisch auswerten
+                 zu können, ermöglicht es, enorm umfassende und gleichzeitig sehr detaillierte Informationen aus
+                 Webdaten gewinnen zu können. In der Wirtschaftsgeographie hat eine Exploration dieser Verfahren bisher
+                 kaum stattgefunden, sodass es das übergeordnete Ziel dieser Disseration ist unstrukturierte Textdaten
+                 aus dem Internet für wirtschaftsgeographische Forschung nutzbar zu machen. Aufgrund des
+                 methodenexplorierenden Charakters der Arbeit führt diese zunächst in die Forschungsfelder Web Mining
+                 und Natural Language Processing ein, bevor die Methodiken anhand von Fallstudien konkret auf
+                 wirtschaftsgeographische Forschungsfragen projiziert werden. Die Fallstudien skizzieren verschiedene
+                 Zugänge zu Webdaten, demonstrieren unterschiedliche Verfahren zur quantitativen Textanalyse, behandeln
+                 Texte unterschiedlicher Sprachen und umfassen sowohl Quer- als auch Längsschnittanalysen. Dabei liegt
+                 der Fokus auf der Entwicklung und Adaptierung von Modellen, die speziell im Kontext raumbezogener
+                 Forschung eingesetzt werden können. Im Rahmen der ersten Fallstudie wurde das offene Webrepositorium
+                 Com-monCrawl genutzt, um eine flächendeckende, koordinatenscharfe Datenbank von Unterneh-mensdomains
+                 mittels Web Mining zu erstellen. Die geographische Analyse und der Vergleich mit amtlichen Statistiken
+                 zeigt, dass die extrahierten Daten in der Lage sind, die tatsächliche Unternehmenslandschaft in
+                 Deutschland zu repräsentieren. Fallstudie 2 nutzt diese Daten, um Unternehmen anhand ihrer
+                 Webseitentexte nach Technologienutzung zu klassifizieren. In der dritten Fallstudie wurde einschlägige
+                 wirtschaftsgeographische Literatur herangezogen, um abstrakte Themen in den Publikationen aufzudecken.
+                 Ferner konnten Entwicklungstrends und Zusammenhänge der Themen mittels Verfahren des Natural Language
+                 Processings quantifiziert werden. Abschließend diskutiert die Arbeit weitere Potentiale und
+                 Herausforderungen der explorierten Methodiken. Die Diskussion beinhaltet ferner eine Gegenüberstellung
+                 der untersuchten Methodiken mit tradierten Verfahren der empirischen Sozialforschung. Aus dieser
+                 Erörterung heraus wurde ebenfalls beleuchtet, wie sich Web Mining und Natural Language Processing
+                 insbesondere in wirtschaftsgeographische Forschungsdesigns integrieren lassen und welche Perspektiven
+                 eine Methodenintegration ermöglicht.",
+  language     = "de",
+  urldate      = "2025-05-16",
+  author       = "Kriesch, Lukas Julian",
+  year         = "2023",
+  cc-author-affiliation = "Justus-Liebig-University Giessen, Germany",
+  cc-class     = "economic geograph, web-mining",
+  cc-dataset-used = "URL index",
+}

bib/cc2024.bib

@@ -340,3 +340,58 @@ @Misc{cc:Amarikwa:2024:Internet-Openness-at-Risk
                  2008, it represents one of the largest publicly accessible web crawl data corpuses on a petabyte scale,
                  and this is one major reason why it’s been used so widely in academia and the industry.”).]",
 }
+
+@Misc{Blagojevic:2024:CC-news-dataset,
+  author       = "Vladimir Blagojevic",
+  year         = "2024",
+  URL          = "https://huggingface.co/datasets/vblagoje/cc_news",
+  cc-derived-dataset-about = "vblagoje-cc-news",
+  cc-author-affiliation = "deepset AI, Germany",
+  cc-snippet   = "CC-News dataset has been proposed, created, and maintained by Sebastian Nagel. The data is publicly
+                 available on AWS S3 Common Crawl bucket at /crawl-data/CC-NEWS/. This version of the dataset has been
+                 prepared using news-please [https://github.com/fhamborg/news-please] - an integrated web crawler and
+                 information extractor for news. It contains 708241 English language news articles published between Jan
+                 2017 and December 2019. Although news-please tags each news article with an appropriate language tag,
+                 these tags are somewhat unreliable. To strictly isolate English language articles an additional check
+                 has been performed using Spacy langdetect pipeline. We selected articles with text fields scores of
+                 80\% probability or more of being English. There are no strict guarantees that each article has all the
+                 relevant fields. For example, 527595 articles have a valid description field. All articles have what
+                 appears to be a valid image URL, but they have not been verified.",
+  cc-class     = "nlp/corpus-construction, news",
+}
+
+@Article{cc:KrieschLosacker:2024:Bioeconomy-firms,
+  place        = "Vienna, Austria",
+  title        = "Bioeconomy firms and where to find them",
+  volume       = "11",
+  URL          = "https://openjournals.wu.ac.at/ojs/index.php/region/article/view/523",
+  doi          = "10.18335/region.v11i1.523",
+  abstractnote = "The bioeconomy represents a transformative approach to economic development and sustainability by
+                 harnessing biological resources and knowledge to produce goods, services, and energy while reducing
+                 dependence on non-renewable resources. In order to understand and support the bioeconomy, scholars and
+                 policymakers rely on an accurate measurement and monitoring of biobased economic activities. However,
+                 existing statistical frameworks and industry classifications often fall short in capturing the unique
+                 characteristics and complexities of the bioeconomy. This article addresses this challenge by developing
+                 a methodological approach for comprehensive measurement and mapping of biobased economic activities. We
+                 build a novel data set of bioeconomy firms in Germany using web-mining and machine learning techniques.
+                 This data set enables detailed analysis of biobased economic activities, providing valuable insights
+                 into the spatial organization of the bioeconomy. The paper demonstrates the applicability of the data
+                 set by testing several stylized facts about the bioeconomy. Our research contributes to a better
+                 understanding of the bioeconomy’s regional impacts and offers a valuable resource for policymakers
+                 and researchers interested in understanding the geography of biobased economic activities. We make an
+                 aggregated version of the data set freely available online.",
+  number       = "1",
+  journal      = "REGION",
+  author       = "Kriesch, Lukas and Losacker, Sebastian",
+  year         = "2024",
+  month        = apr,
+  pages        = "55–78",
+  cc-author-affiliation = "Justus-Liebig-University Giessen, Germany",
+  cc-class     = "economic geography, web-mining, regional science, regional economics",
+  cc-snippet   = "The dataset is based on a novel web-mining approach developed by Kriesch (2023). This dataset uses the
+                 open-source web repository CommonCrawl to identify German company websites and has proven to be a
+                 valuable database for spatial research. From this data, we identify bioeconomy firms using a
+                 combination of different natural language processing techniques, utilizing the semantic capabilities of
+                 modern transformer models (Reimers, Gurevych 2019, Vaswani et al. 2017).",
+  cc-dataset-used = "URL index",
+}