2025
Bachelor Theses
1.
Delarue, Simon
Learning on graphs : from algorithms to socio-technical analyses on AI Bachelor Thesis
Telecom Paris, 2025, (HAL Id: tel-04963643).
@bachelorthesis{Delarue2025,
title = {Learning on graphs : from algorithms to socio-technical analyses on AI},
author = {Simon Delarue},
url = {https://theses.hal.science/tel-04963643v1/file/142478_DELARUE_2025_archivage.pdf
https://theses.hal.science/tel-04963643v1},
year = {2025},
date = {2025-02-24},
school = {Telecom Paris},
abstract = {Artificial Intelligence (AI) techniques and algorithms are now deeply embedded in our daily lives. While they offer promising possibilities, their expanding presence in high- stake decision-making domains such as healthcare, justice, and industry raises significant societal, ethical, environmental, and governance issues. Far from being neutral tools, AI models have transformed our practices by introducing or amplifying biases – such as those related to race or gender – and have led to broader consequences, reshaping debates on issues such as the environmental impact of digitalisation. To address the challenges posed by the rapid expansion of AI technologies, and as these issues begin to be considered within international regulations like the AI Act in Europe, it is essential for scientists and developers to incorporate these considerations into their approaches. This requires moving beyond a restricted focus on performance to include aspects like scalability, simplicity, and explainability. This thesis examines the potential of attributed graph-based approaches to address both the technical challenges posed by AI methods, and the socio-technical questions that arise from their complex relationship with society. The first part of this thesis explores how graph-based methods can meet the require- ments for efficiency, scalability, and simplicity in learning techniques. Graphs, i.e. sets of nodes connected to each other through edges, enable the modelling of complex re- lational data and draw on contributions from fields ranging from computer science to social sciences, offering promising solutions to the limitations encountered in AI. First, through a software contribution, we show how the inherent sparsity of complex networks can be leveraged within model implementation to reduce the computational cost of cur- rent approaches. Then, by examining the capabilities of non-neural attributed graph approaches, this thesis shows that simple methods can outperform state-of-the-art neu- ral networks in capturing the structural complexity of real-world data, thereby providing scalable and generalisable solutions to graph-based machine learning tasks such as node classification and link prediction. Finally, we design an attributed pattern mining ap- proach to derive interesting and easily understandable insights from complex networks. Recognising the need for diverse analytical approaches to understand the complex entanglements among current AI techniques, what we aim to achieve with them, and how they transform our uses, the second part of this thesis shifts to an interdisciplinary exam- ination of AI as a socio-technical system. This part explores how AI can be re-envisioned not only as a tool but also as an object of study. By framing AI as an ecosystem shaped by diverse stakeholders and societal concerns, this thesis uses graph-based models to map the interactions and tensions within AI, particularly around explainability, ethics, and environmental impact. For this purpose, we conduct a user study to examine the po- v tential of attributed graph-based explanations to enhance users’ perception of AI-based recommendations, and reveal the complex link between users’ preference towards expla- nation design and the understanding gain these explanations allow. In a second analysis, we build a corpus of AI charters and manifestos for ethics, which we make publicly avail- able. Using this corpus, we quantitatively analyse the interactions among key actors forming the social world of AI ethics in order to understand their roles in influencing AI governance and regulation. Finally, we explore how AI-related scientists incorporate their environmental concerns into their research using attributed graph analysis. This study reveals that environmental concerns remain largely framed through a technical perspective, with little consideration of the ecological impacts of digitalisation, pointing to the need for a more balanced approach in future research on AI and the environment. Building directly on the work presented in this thesis, we conclude by opening path- ways for future research directions focused on simple and efficient graph-based approaches to learning on complex networks. In a broader context, we also discuss future research avenues that we consider essential, including research rooted in Science and Technology Studies, to assess how digital technologies might evolve as inclusive, responsible, and sustainable tools. },
note = {HAL Id: tel-04963643},
keywords = {},
pubstate = {published},
tppubtype = {bachelorthesis}
}
Artificial Intelligence (AI) techniques and algorithms are now deeply embedded in our daily lives. While they offer promising possibilities, their expanding presence in high- stake decision-making domains such as healthcare, justice, and industry raises significant societal, ethical, environmental, and governance issues. Far from being neutral tools, AI models have transformed our practices by introducing or amplifying biases – such as those related to race or gender – and have led to broader consequences, reshaping debates on issues such as the environmental impact of digitalisation. To address the challenges posed by the rapid expansion of AI technologies, and as these issues begin to be considered within international regulations like the AI Act in Europe, it is essential for scientists and developers to incorporate these considerations into their approaches. This requires moving beyond a restricted focus on performance to include aspects like scalability, simplicity, and explainability. This thesis examines the potential of attributed graph-based approaches to address both the technical challenges posed by AI methods, and the socio-technical questions that arise from their complex relationship with society. The first part of this thesis explores how graph-based methods can meet the require- ments for efficiency, scalability, and simplicity in learning techniques. Graphs, i.e. sets of nodes connected to each other through edges, enable the modelling of complex re- lational data and draw on contributions from fields ranging from computer science to social sciences, offering promising solutions to the limitations encountered in AI. First, through a software contribution, we show how the inherent sparsity of complex networks can be leveraged within model implementation to reduce the computational cost of cur- rent approaches. Then, by examining the capabilities of non-neural attributed graph approaches, this thesis shows that simple methods can outperform state-of-the-art neu- ral networks in capturing the structural complexity of real-world data, thereby providing scalable and generalisable solutions to graph-based machine learning tasks such as node classification and link prediction. Finally, we design an attributed pattern mining ap- proach to derive interesting and easily understandable insights from complex networks. Recognising the need for diverse analytical approaches to understand the complex entanglements among current AI techniques, what we aim to achieve with them, and how they transform our uses, the second part of this thesis shifts to an interdisciplinary exam- ination of AI as a socio-technical system. This part explores how AI can be re-envisioned not only as a tool but also as an object of study. By framing AI as an ecosystem shaped by diverse stakeholders and societal concerns, this thesis uses graph-based models to map the interactions and tensions within AI, particularly around explainability, ethics, and environmental impact. For this purpose, we conduct a user study to examine the po- v tential of attributed graph-based explanations to enhance users’ perception of AI-based recommendations, and reveal the complex link between users’ preference towards expla- nation design and the understanding gain these explanations allow. In a second analysis, we build a corpus of AI charters and manifestos for ethics, which we make publicly avail- able. Using this corpus, we quantitatively analyse the interactions among key actors forming the social world of AI ethics in order to understand their roles in influencing AI governance and regulation. Finally, we explore how AI-related scientists incorporate their environmental concerns into their research using attributed graph analysis. This study reveals that environmental concerns remain largely framed through a technical perspective, with little consideration of the ecological impacts of digitalisation, pointing to the need for a more balanced approach in future research on AI and the environment. Building directly on the work presented in this thesis, we conclude by opening path- ways for future research directions focused on simple and efficient graph-based approaches to learning on complex networks. In a broader context, we also discuss future research avenues that we consider essential, including research rooted in Science and Technology Studies, to assess how digital technologies might evolve as inclusive, responsible, and sustainable tools.
2023
Proceedings Articles
2.
Berrou, Yolène; Soulier, Eddie
A Methodology to Analyze the Development of Local Energy Communities Based on Socio-Energetic Nodes and Actor-Network Theory Proceedings Article
In: pp. 439-446, Elsevier, 2023, ISSN: 1877-0509, (CENTERIS – International Conference on ENTERprise Information Systems / ProjMAN – International Conference on Project MANagement / HCist – International Conference on Health and Social Care Information Systems and Technologies 2022).
@inproceedings{Berrou2023,
title = {A Methodology to Analyze the Development of Local Energy Communities Based on Socio-Energetic Nodes and Actor-Network Theory},
author = {Yolène Berrou and Eddie Soulier},
url = {https://www.sciencedirect.com/science/article/pii/S1877050923003198},
doi = {/10.1016/j.procs.2023.01.310},
issn = {1877-0509},
year = {2023},
date = {2023-03-22},
urldate = {2023-03-22},
journal = {Procedia Computer Science},
volume = {219},
pages = {439-446},
publisher = {Elsevier},
abstract = {The shift from centralized to decentralized energy, with the development of renewable energies, is giving rise to new energy models. Some of these models aim to increase the citizens participation in the energy transition, such as the energy communities. This concept has recently emerged in Europe to encourage the development of local projects and raising citizens' awareness. Our aim is to better understand how such communities emerge to foster them, and to propose a tool for B2T (Business to Territory) Business Developers. We have developed a generic methodology to follow the formation of sociotechnical systems based on a modeling of the Actor-Network Theory. We use the concept of Socio-Energetic Node and propose a model of it to apply our generic methodology to Local Energy Communities. Preliminary results are presented at the end of this paper on a case study.},
note = {CENTERIS – International Conference on ENTERprise Information Systems / ProjMAN – International Conference on Project MANagement / HCist – International Conference on Health and Social Care Information Systems and Technologies 2022},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
The shift from centralized to decentralized energy, with the development of renewable energies, is giving rise to new energy models. Some of these models aim to increase the citizens participation in the energy transition, such as the energy communities. This concept has recently emerged in Europe to encourage the development of local projects and raising citizens' awareness. Our aim is to better understand how such communities emerge to foster them, and to propose a tool for B2T (Business to Territory) Business Developers. We have developed a generic methodology to follow the formation of sociotechnical systems based on a modeling of the Actor-Network Theory. We use the concept of Socio-Energetic Node and propose a model of it to apply our generic methodology to Local Energy Communities. Preliminary results are presented at the end of this paper on a case study.
Masters Theses
3.
Zafar, Lubna
Impact of Field of Study Trend on Citation Count of Scientific Articles and Authors Masters Thesis
The Capital University of Science and Technology, Islamabad, Pakistan, 2023.
@mastersthesis{Zafar2023,
title = {Impact of Field of Study Trend on Citation Count of Scientific Articles and Authors},
author = {Lubna Zafar},
url = {https://cust.edu.pk/wp-content/uploads/2023/10/Dissertation-Lubna-Zafar.pdf
},
year = {2023},
date = {2023-07-17},
address = {Islamabad, Pakistan},
school = {The Capital University of Science and Technology},
abstract = {Millions of new scientific articles are published each year. Researchers work and publish in their respective fields of interest. A major portion of the scientific community publishing in the same Field of Study (FoS) forms a trend of that field. A novice researcher chooses his field of interest based upon its popularity.
This may have a positive impact on the acceptance of a study or high count of citations in future. There are multiple studies in literature that focus on FoS trend detection and analysis, birth and establishment of an FoS trend, number of publications and researchers in an FoS trend, communities of researchers being formed around an FoS trend, author’s FoS switching, vanishing of an FoS trend, trends in different disciplines etc. However, the previous work contains a gap, that is, there is no work on impact of following an FoS on citation trend of scientific articles and authors.
This study identifies how significant it is to follow an FoS trend and the impact of the FoS trend on research paper citations and on authors citation count. For this purpose, we have chosen the field of Computer Science and Microsoft Academic Graph (MAG) dataset from the 1950-2018 time period. We extracted publications of different FoS of Computer Science and also citation counts for these publications.
First, we established similarity between citation trends of papers belonging to same FoS using rand index and correlation. Then we proposed a technique to identify trend setters and trend followers that would help to identify influential authors in a particular FoS. Finally, we established the impact of FoS on the citation patterns of authors by achieving a consistent R2 values of papers belonging to same FoS.
The results depict that if papers belong to the same FoS, then there are 69% of the chances of having a similar citation pattern and that they have the same citation trend as they also have achieved a high correlation value. Experimental results show that there is a similarity between citation trend of authors that belong to the same FoS as compared to different FoS and achieved consistent R2 value. FoS trend following has a certain impact on the citation count of authors. The result also shows that if an author publishes in a particular FoS, then the citation trend of this author’s work resembles more to the overall citation trend of that particular FoS than that of some other FoS. This proves that FoS has a certain impact on the citation count of a paper and researchers should contemplate on the FoS trend before selecting a particular research area
},
keywords = {},
pubstate = {published},
tppubtype = {mastersthesis}
}
Millions of new scientific articles are published each year. Researchers work and publish in their respective fields of interest. A major portion of the scientific community publishing in the same Field of Study (FoS) forms a trend of that field. A novice researcher chooses his field of interest based upon its popularity.
This may have a positive impact on the acceptance of a study or high count of citations in future. There are multiple studies in literature that focus on FoS trend detection and analysis, birth and establishment of an FoS trend, number of publications and researchers in an FoS trend, communities of researchers being formed around an FoS trend, author’s FoS switching, vanishing of an FoS trend, trends in different disciplines etc. However, the previous work contains a gap, that is, there is no work on impact of following an FoS on citation trend of scientific articles and authors.
This study identifies how significant it is to follow an FoS trend and the impact of the FoS trend on research paper citations and on authors citation count. For this purpose, we have chosen the field of Computer Science and Microsoft Academic Graph (MAG) dataset from the 1950-2018 time period. We extracted publications of different FoS of Computer Science and also citation counts for these publications.
First, we established similarity between citation trends of papers belonging to same FoS using rand index and correlation. Then we proposed a technique to identify trend setters and trend followers that would help to identify influential authors in a particular FoS. Finally, we established the impact of FoS on the citation patterns of authors by achieving a consistent R2 values of papers belonging to same FoS.
The results depict that if papers belong to the same FoS, then there are 69% of the chances of having a similar citation pattern and that they have the same citation trend as they also have achieved a high correlation value. Experimental results show that there is a similarity between citation trend of authors that belong to the same FoS as compared to different FoS and achieved consistent R2 value. FoS trend following has a certain impact on the citation count of authors. The result also shows that if an author publishes in a particular FoS, then the citation trend of this author’s work resembles more to the overall citation trend of that particular FoS than that of some other FoS. This proves that FoS has a certain impact on the citation count of a paper and researchers should contemplate on the FoS trend before selecting a particular research area
This may have a positive impact on the acceptance of a study or high count of citations in future. There are multiple studies in literature that focus on FoS trend detection and analysis, birth and establishment of an FoS trend, number of publications and researchers in an FoS trend, communities of researchers being formed around an FoS trend, author’s FoS switching, vanishing of an FoS trend, trends in different disciplines etc. However, the previous work contains a gap, that is, there is no work on impact of following an FoS on citation trend of scientific articles and authors.
This study identifies how significant it is to follow an FoS trend and the impact of the FoS trend on research paper citations and on authors citation count. For this purpose, we have chosen the field of Computer Science and Microsoft Academic Graph (MAG) dataset from the 1950-2018 time period. We extracted publications of different FoS of Computer Science and also citation counts for these publications.
First, we established similarity between citation trends of papers belonging to same FoS using rand index and correlation. Then we proposed a technique to identify trend setters and trend followers that would help to identify influential authors in a particular FoS. Finally, we established the impact of FoS on the citation patterns of authors by achieving a consistent R2 values of papers belonging to same FoS.
The results depict that if papers belong to the same FoS, then there are 69% of the chances of having a similar citation pattern and that they have the same citation trend as they also have achieved a high correlation value. Experimental results show that there is a similarity between citation trend of authors that belong to the same FoS as compared to different FoS and achieved consistent R2 value. FoS trend following has a certain impact on the citation count of authors. The result also shows that if an author publishes in a particular FoS, then the citation trend of this author’s work resembles more to the overall citation trend of that particular FoS than that of some other FoS. This proves that FoS has a certain impact on the citation count of a paper and researchers should contemplate on the FoS trend before selecting a particular research area
2021
Journal Articles
4.
León-Vargas, Fabian; Oviedo, Jineth Andrea Arango; Wandurraga, Héctor Javier Luna
Two Decades of Research in Artificial Pancreas: Insights from a Bibliometric Analysis Journal Article
In: Journal of Diabetes Science and Technology, vol. 0, no. 0, pp. 19322968211005500, 2021.
@article{León-Vargas2021,
title = {Two Decades of Research in Artificial Pancreas: Insights from a Bibliometric Analysis},
author = {Fabian León-Vargas and Jineth Andrea Arango Oviedo and Héctor Javier Luna Wandurraga},
url = {https://doi.org/10.1177/19322968211005500},
doi = {10.1177/19322968211005500},
year = {2021},
date = {2021-04-15},
urldate = {2021-04-15},
journal = {Journal of Diabetes Science and Technology},
volume = {0},
number = {0},
pages = {19322968211005500},
abstract = {Artificial pancreas is a well-known research topic devoted to achieving better glycemic outcomes that has been attracting increasing attention over the years. However, there is a lack of systematic, chronological, and synthesizing studies that show the background of the knowledge generation in this field. This study implements a bibliometric analysis to recognize the main documents, type of publications, research categories, countries, keywords, organizations, and authors related to this topic.Methods:Web of Science core collection database was accessed from 2000 to 2020 in order to select high-quality scientific documents based on a specific search query. Bibexcel, MS Excel, Power BI, R-Studio, VOSviewer, and CorText software were used for a descriptive and network analysis based on the local database obtained. Bibliometric parameters as the h-index, frequencies, co-authorship and co-ocurrences were computed.Results:A total of 756 documents were included that show a growing scientific production on this topic with an increasing contribution from engineering. Outstanding authors, organizations, and countries were identified. An analysis of trends in research was conducted according to the scientific categories of the Web of Science database to identify the main research interests of the last 2 decades and the emerging areas with greater prominence in the coming years. A keyword network analysis allowed to identify the main stages in the development of the AP research over time.Conclusions:Results reveal a comprehensive background of the knowledge generation for the AP topic during the last 2 decades, which has been strengthened with international collaborations and a remarkable interdisciplinarity between endocrinology and engineering, giving rise to a growing number of research areas over time, where computer science and medical informatics stand out as the main emerging research areas.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Artificial pancreas is a well-known research topic devoted to achieving better glycemic outcomes that has been attracting increasing attention over the years. However, there is a lack of systematic, chronological, and synthesizing studies that show the background of the knowledge generation in this field. This study implements a bibliometric analysis to recognize the main documents, type of publications, research categories, countries, keywords, organizations, and authors related to this topic.Methods:Web of Science core collection database was accessed from 2000 to 2020 in order to select high-quality scientific documents based on a specific search query. Bibexcel, MS Excel, Power BI, R-Studio, VOSviewer, and CorText software were used for a descriptive and network analysis based on the local database obtained. Bibliometric parameters as the h-index, frequencies, co-authorship and co-ocurrences were computed.Results:A total of 756 documents were included that show a growing scientific production on this topic with an increasing contribution from engineering. Outstanding authors, organizations, and countries were identified. An analysis of trends in research was conducted according to the scientific categories of the Web of Science database to identify the main research interests of the last 2 decades and the emerging areas with greater prominence in the coming years. A keyword network analysis allowed to identify the main stages in the development of the AP research over time.Conclusions:Results reveal a comprehensive background of the knowledge generation for the AP topic during the last 2 decades, which has been strengthened with international collaborations and a remarkable interdisciplinarity between endocrinology and engineering, giving rise to a growing number of research areas over time, where computer science and medical informatics stand out as the main emerging research areas.
2020
Journal Articles
5.
Deng, Shengli; Xia, Sudi
Mapping the interdisciplinarity in information behavior research: a quantitative study using diversity measure and co‐occurrence analysis Journal Article
In: 2020.
@article{Deng2020b,
title = {Mapping the interdisciplinarity in information behavior research: a quantitative study using diversity measure and co‐occurrence analysis},
author = {Shengli Deng and Sudi Xia},
doi = {https://doi.org/10.1007/s11192-020-03465-x},
year = {2020},
date = {2020-04-11},
urldate = {2020-04-11},
abstract = {Information behavior research is an interdisciplinary field in essence due to the investiga- tion of interdisciplinary in previous work. To track the changes in interdisciplinarity of this field, more efforts should be put on basis of previous work. Based on publications searched from Web of Science from 2000 to 2018, we explored the interdisciplinarity of this field drawing on network analysis and diversity measure. Findings showed that although variety of disciplines in this field augmented significantly, the distribution of disciplines is unbal- anced and concentrated on some dominant disciplines such as computer science, engineer- ing, psychology, social science and medicine, etc. Relationships among disciplines have evolved over time and mainly focused on neighboring disciplines instead of distinct disci- plines. Computer science, engineering, psychology, health science and social science func- tion as intermediate disciplines connecting distinct disciplinary groups. Besides, the meas- urement using diversity measure shows that interdisciplinary degree of this field appears to decrease. This study contributes to the evolution and measurement of interdisciplinarity of information behavior research, which has implications for researchers and practitioners in this field.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Information behavior research is an interdisciplinary field in essence due to the investiga- tion of interdisciplinary in previous work. To track the changes in interdisciplinarity of this field, more efforts should be put on basis of previous work. Based on publications searched from Web of Science from 2000 to 2018, we explored the interdisciplinarity of this field drawing on network analysis and diversity measure. Findings showed that although variety of disciplines in this field augmented significantly, the distribution of disciplines is unbal- anced and concentrated on some dominant disciplines such as computer science, engineer- ing, psychology, social science and medicine, etc. Relationships among disciplines have evolved over time and mainly focused on neighboring disciplines instead of distinct disci- plines. Computer science, engineering, psychology, health science and social science func- tion as intermediate disciplines connecting distinct disciplinary groups. Besides, the meas- urement using diversity measure shows that interdisciplinary degree of this field appears to decrease. This study contributes to the evolution and measurement of interdisciplinarity of information behavior research, which has implications for researchers and practitioners in this field.
2018
Technical Reports
6.
Aubin, Sophie; Huber, Madeleine
eROSA. e-infrastructure Roadmap for Open Science in Agriculture : bibliometric study results Technical Report
INRA, Horizon 2020 2018, (Ref. Ares(2018)3404573).
@techreport{Aubin2018,
title = {eROSA. e-infrastructure Roadmap for Open Science in Agriculture : bibliometric study results},
author = {Sophie Aubin and Madeleine Huber},
url = {https://zenodo.org/record/1305000/files/Bibliometric%20study%20results.pdf},
year = {2018},
date = {2018-06-28},
urldate = {2018-06-28},
institution = {INRA, Horizon 2020},
abstract = {This study highlights the results of a bibliometric analysis conducted at a global scale in order to identify key scientists and associated research performing organisations (e.g. public research institutes, universities, Research & Development departments of private companies) that work in the field of agricultural data sources and services. The added value of such a methodological approach is the resulting ability to provide a detailed answer to the question “who does what?” by collecting, processing, analysing and visualising the metadata1 of related scientific publications. The study focuses on articles that have been published in the past 10 years (i.e. during the period 2005-2015). As such, the analysis is a first attempt at delineating, mapping and describing the scientific community that the e-ROSA project seeks to engage with. It neither aims at being exhaustive nor at providing an evaluation on the scientific excellence of identified stakeholders as this is not the goal of the community-building activity under e-ROSA. The specific objectives of the analysis include:
1 - The identification of scientists and related collaboration networks involved in data science for agriculture in order to initiate further contact while building and engaging with the e-ROSA community throughout the project: e.g. these results provide valuable contacts in the context of the desk surveys that will be carried out under Work Package 1 in order to consolidate and reach out to the community, and in the context of the workshops organised under Work Package 2 that seek community-building and co-design of the e-ROSA Roadmap.
2 - The identification of specific domains related to data and computer science that are of interest to identified scientists (i.e. working on agricultural issues).
3 - The identification of related conferences and journals that the e-ROSA project can target in order to effectively reach out to the relevant communities involved in data science issues related to agriculture.},
note = {Ref. Ares(2018)3404573},
keywords = {},
pubstate = {published},
tppubtype = {techreport}
}
This study highlights the results of a bibliometric analysis conducted at a global scale in order to identify key scientists and associated research performing organisations (e.g. public research institutes, universities, Research & Development departments of private companies) that work in the field of agricultural data sources and services. The added value of such a methodological approach is the resulting ability to provide a detailed answer to the question “who does what?” by collecting, processing, analysing and visualising the metadata1 of related scientific publications. The study focuses on articles that have been published in the past 10 years (i.e. during the period 2005-2015). As such, the analysis is a first attempt at delineating, mapping and describing the scientific community that the e-ROSA project seeks to engage with. It neither aims at being exhaustive nor at providing an evaluation on the scientific excellence of identified stakeholders as this is not the goal of the community-building activity under e-ROSA. The specific objectives of the analysis include:
1 - The identification of scientists and related collaboration networks involved in data science for agriculture in order to initiate further contact while building and engaging with the e-ROSA community throughout the project: e.g. these results provide valuable contacts in the context of the desk surveys that will be carried out under Work Package 1 in order to consolidate and reach out to the community, and in the context of the workshops organised under Work Package 2 that seek community-building and co-design of the e-ROSA Roadmap.
2 - The identification of specific domains related to data and computer science that are of interest to identified scientists (i.e. working on agricultural issues).
3 - The identification of related conferences and journals that the e-ROSA project can target in order to effectively reach out to the relevant communities involved in data science issues related to agriculture.
1 - The identification of scientists and related collaboration networks involved in data science for agriculture in order to initiate further contact while building and engaging with the e-ROSA community throughout the project: e.g. these results provide valuable contacts in the context of the desk surveys that will be carried out under Work Package 1 in order to consolidate and reach out to the community, and in the context of the workshops organised under Work Package 2 that seek community-building and co-design of the e-ROSA Roadmap.
2 - The identification of specific domains related to data and computer science that are of interest to identified scientists (i.e. working on agricultural issues).
3 - The identification of related conferences and journals that the e-ROSA project can target in order to effectively reach out to the relevant communities involved in data science issues related to agriculture.
2017
Journal Articles
7.
Hu, Jiming; Zhang, Yin
Discovering the interdisciplinary nature of Big Data research through social network analysis and visualization Journal Article
In: Scientometrics, vol. 112, pp. 91–109, 2017.
@article{Hu2017,
title = {Discovering the interdisciplinary nature of Big Data research through social network analysis and visualization},
author = {Jiming Hu and Yin Zhang},
doi = {10.1007/s11192-017-2383-1},
year = {2017},
date = {2017-05-01},
urldate = {2017-05-01},
journal = {Scientometrics},
volume = {112},
pages = {91–109},
abstract = {Big Data is a research field involving a large number of collaborating disciplines. Based on bibliometric data downloaded from the Web of Science, this study applies various social network analysis and visualization tools to examine the structure and patterns of interdisciplinary collaborations, as well as the recently evolving overall pattern. This study presents the descriptive statistics of disciplines involved in publishing Big Data research; and network indicators of the interdisciplinary collaborations among disciplines, interdisciplinary communities, interdisciplinary networks, and changes in discipline communities over time. The findings indicate that the scope of disciplines involved in Big Data research is broad, but that the disciplinary distribution is unbalanced. The overall collaboration among disciplines tends to be concentrated in several key fields. According to the network indicators, Computer Science, Engineering, and Business and Economics are the most important contributors to Big Data research, given their position and role in the research collaboration network. Centering around a few important disciplines, all fields related to Big Data research are aggregated into communities, suggesting some related research areas, and directions for Big Data research. An ever-changing roster of related disciplines provides support, as illustrated by the evolving graph of communities.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Big Data is a research field involving a large number of collaborating disciplines. Based on bibliometric data downloaded from the Web of Science, this study applies various social network analysis and visualization tools to examine the structure and patterns of interdisciplinary collaborations, as well as the recently evolving overall pattern. This study presents the descriptive statistics of disciplines involved in publishing Big Data research; and network indicators of the interdisciplinary collaborations among disciplines, interdisciplinary communities, interdisciplinary networks, and changes in discipline communities over time. The findings indicate that the scope of disciplines involved in Big Data research is broad, but that the disciplinary distribution is unbalanced. The overall collaboration among disciplines tends to be concentrated in several key fields. According to the network indicators, Computer Science, Engineering, and Business and Economics are the most important contributors to Big Data research, given their position and role in the research collaboration network. Centering around a few important disciplines, all fields related to Big Data research are aggregated into communities, suggesting some related research areas, and directions for Big Data research. An ever-changing roster of related disciplines provides support, as illustrated by the evolving graph of communities.
Online
8.
Salatino, Angelo A; Osborne, Francesco; Motta, Enrico
How are topics born? Understanding the research dynamics preceding the emergence of new areas Online
2017, visited: 01.01.2017.
@online{Salatino2016,
title = {How are topics born? Understanding the research dynamics preceding the emergence of new areas},
author = {Angelo A Salatino and Francesco Osborne and Enrico Motta},
url = {https://doi.org/10.7717/peerj-cs.119},
doi = {10.7717/peerj-cs.119},
year = {2017},
date = {2017-01-01},
urldate = {2017-01-01},
journal = {PeerJ Computer Science},
abstract = {The ability to promptly recognise new research trends is strategic for many stakeholders, including universities, institutional funding bodies, academic publishers and companies. While the literature describes several approaches which aim to identify the emergence of new research topics early in their lifecycle, these rely on the assumption that the topic in question is already associated with a number of publications and consistently referred to by a community of researchers. Hence, detecting the emergence of a new research area at an embryonic stage, i.e., before the topic has been consistently labelled by a community of researchers and associated with a number of publications, is still an open challenge. In this paper, we begin to address this challenge by performing a study of the dynamics preceding the creation of new topics. This study indicates that the emergence of a new topic is anticipated by a significant increase in the pace of collaboration between relevant research areas, which can be seen as the ‘parents’ of the new topic. These initial findings (i) confirm our hypothesis that it is possible in principle to detect the emergence of a new topic at the embryonic stage, (ii) provide new empirical evidence supporting relevant theories in Philosophy of Science, and also (iii) suggest that new topics tend to emerge in an environment in which weakly interconnected research areas begin to cross-fertilise.},
keywords = {},
pubstate = {published},
tppubtype = {online}
}
The ability to promptly recognise new research trends is strategic for many stakeholders, including universities, institutional funding bodies, academic publishers and companies. While the literature describes several approaches which aim to identify the emergence of new research topics early in their lifecycle, these rely on the assumption that the topic in question is already associated with a number of publications and consistently referred to by a community of researchers. Hence, detecting the emergence of a new research area at an embryonic stage, i.e., before the topic has been consistently labelled by a community of researchers and associated with a number of publications, is still an open challenge. In this paper, we begin to address this challenge by performing a study of the dynamics preceding the creation of new topics. This study indicates that the emergence of a new topic is anticipated by a significant increase in the pace of collaboration between relevant research areas, which can be seen as the ‘parents’ of the new topic. These initial findings (i) confirm our hypothesis that it is possible in principle to detect the emergence of a new topic at the embryonic stage, (ii) provide new empirical evidence supporting relevant theories in Philosophy of Science, and also (iii) suggest that new topics tend to emerge in an environment in which weakly interconnected research areas begin to cross-fertilise.
Workshops
9.
Cointet, Jean-Philippe; Abdo, Alexandre Hannud
Capturing Oncology Dynamics from Textual Content of Conference Abstracts: Word Embedding and Stochastic Block Models Workshop
2017.
@workshop{cointet2017capturing,
title = {Capturing Oncology Dynamics from Textual Content of Conference Abstracts: Word Embedding and Stochastic Block Models},
author = {Jean-Philippe Cointet and Alexandre Hannud Abdo},
url = {http://www.ixxi.fr/agenda/seminaires/understanding-the-dynamics-of-science-an-interdisciplinary-workshop?searchterm=Understanding+the+dynamics+of+science},
year = {2017},
date = {2017-01-01},
urldate = {2017-01-01},
abstract = {The availability of social data drives many scientists from the formal sciences (computer science, physics…) into the quantitative analysis of social systems. One early example of this trend is « scientometrics », the study of science’s structure and evolutions using large bibliographic datasets. Recent topics of interest in the field include the development of new formal tools to provide insights on the nature, structure and dynamics of scientific communities « bottom-up », i.e. without using predetermined classification schemes. Many scientists develop also interactive visualization platforms, or compare the pictures obtained by quantitative and qualitative methods.},
keywords = {},
pubstate = {published},
tppubtype = {workshop}
}
The availability of social data drives many scientists from the formal sciences (computer science, physics…) into the quantitative analysis of social systems. One early example of this trend is « scientometrics », the study of science’s structure and evolutions using large bibliographic datasets. Recent topics of interest in the field include the development of new formal tools to provide insights on the nature, structure and dynamics of scientific communities « bottom-up », i.e. without using predetermined classification schemes. Many scientists develop also interactive visualization platforms, or compare the pictures obtained by quantitative and qualitative methods.
LIST OF SCIENTIFIC WORKS THAT HAVE USED CORTEXT MANAGER
(Sources: Google Scholar, HAL, Scopus, WOS and search engines)
We are grateful that you have found CorTexT Manager useful. Over the years, you have been more than 1050 authors to trust CorTexT for your publicly accessible analyzes. This represents a little less than 10% of CorTexT Manager user’s community. So, thank you!
We seek to understand how the scientific production that used CorText Manager has evolved and to characterise it. You will find here our analysis of this scientific production.
Browse documents by main topics
| What types of documents? |
|---|
| What types of documents? |
| 254 journal articles |
| 46 Ph.D. thesis |
| 43 conference (not in proceedings) |
| 43 conference proceedings |
| 32 online articles |
| 31 reports |
| 29 bachelorthesis |
| 26 book chapters |
| 22 masters thesis |
| 14 book |
| 13 workshop |
| 5 miscellaneous |
| 4 proceedings |
| 3 presentation |
| 2 workingpaper |
| 1 manual |
| 1 unpublished |
| Main peer-reviewed journals |
|---|
| Main peer-reviewed journals |
| Scientometrics |
| I2D - Information, données & documents |
| Réseaux |
| Revue d’anthropologie des connaissances |
| PloS one |
| Journal of Rural Studies |
| Library Hi Tech |
| Revue d'anthropologie des connaissances |
| médecine/sciences |
| Renewable Energy |
Browse narrower topics
