Finding cohesive clusters for analyzing knowledge communities

Kandylas, Vasileios; Upham, S. Phineas; Ungar, Lyle H.

doi:10.1007/s10115-008-0135-5

Cited by 22 publications

(10 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…), leading to the development of many techniques that fall outside the classics studied in statistical textbooks. Examples of this are the simple‐centers algorithm (Cobo, López‐Herrera, Herrera‐Viedma, & Herrera, ; Coulter, Monarch, & Konda, ; Muñoz‐Leiva, Viedma‐del‐Jesús, Sánchez‐Fernández, & López‐Herrera, ), streamer (Kandylas, Upham, & Ungar, ), and spectral clustering (Chen, Ibekwe‐SanJuan, & Hou, ), among others (Callon, Courtial, & Laville, ). Classical clustering techniques also have a solid background in bibliometrics, as is the case with hierarchical clustering (Kopcsa & Schiebel, ; Leydesdorff, ; McCain, ; Zong et al., ) and k ‐means clustering (Bassecoulard, Lelu, & Zitt, ; Chang, ; Janssens, Leta, Glänzel, & De Moor, ).…”

Section: Experimental Methodsmentioning

confidence: 99%

Clusterization and mapping of waste recycling science. Evolution of research from 2002 to 2012

Garechana

Belver

Carrasco

et al. 2014

Asso for Info Science & Tech

View full text Add to dashboard Cite

Characterization of waste recycling (WR) research has to start by defining the scope of this scientific area. Previous works and expert assessment recommend the adoption of an inclusive definition, with the aim of including all relevant uses of waste within this study. An ad‐hoc designed “capture” strategy has been used to retrieve WR‐related peer‐reviewed journal papers from selected databases, and the information contained in their author keyword field has been thoroughly cleaned. Author keyword co‐occurrence data have been used for building a similarity measure between keywords, plus cluster analysis for revealing the main WR research being addressed by the scientific community. Results have been further analyzed using advanced visualization tools to determine which clusters formed strongly‐linked research areas that could set the main cognitive divisions of WR science. This process has been repeated with 2002 and 2012 data, and science maps reflecting the main research areas and clusters have been generated. Results show that WR mainly deals with the recovery of basic, widely used raw materials like water and fertile soil. Energy generation and waste management are other relevant fields that show an interesting evolution, revealing signs of growth in research, together with the emergence of sub‐areas reflecting consolidating research specialties.

show abstract

Section: Experimental Methodsmentioning

confidence: 99%

Clusterization and mapping of waste recycling science. Evolution of research from 2002 to 2012

Garechana

Belver

Carrasco

et al. 2014

Asso for Info Science & Tech

View full text Add to dashboard Cite

show abstract

“…Part of our work focuses on applying our evaluation techniques on citations graphs (DBLP, ArXiv). Recent work on citation graphs can be found in [4] where a study is carried out on the citation graph of Computer Science Literature and [31]. In [4], an attempt is made to extract a descriptive summary of the graph through a study of fundamental and well-established properties (degree distribution, giant component size etc.).…”

Section: Related Workmentioning

confidence: 99%

“…In contrast, our work focuses on novel techniques for evaluating community graphs and expands on a wider scope of study. In [31], the focus is on community detection and the evolution through time. The community detection is performed on the authors through the papers they have co-cited, and the evaluation of the citation graph is based on the detected clusters.…”

Section: Related Workmentioning

confidence: 99%

D-cores: measuring collaboration of directed graphs based on degeneracy

2012

View full text Add to dashboard Cite

International audienceCommunity detection and evaluation is an important task in graph mining. In many cases, a community is defined as a subgraph characterized by dense connections or interactions between its nodes. A variety of measures are proposed to evaluate different quality aspects of such communities--in most cases ignoring the directed nature of edges. In this paper, we introduce novel metrics for evaluating the collaborative nature of directed graphs--a property not captured by the single node metrics or by other established commu- nity evaluation metrics. In order to accomplish this objective, we capitalize on the concept of graph degeneracy and define a novel D-core framework, extending the classic graph-theoretic notion of k-cores for undirected graphs to directed ones. Based on the D-core, which essen- tially can be seen as a measure of the robustness of a community under degeneracy, we devise a wealth of novel metrics used to evaluate graph collaboration features of directed graphs. We applied the D-core approach on large synthetic and real-world graphs such as Wikipedia, DBLP, and ArXiv and report interesting results at the graph as well at the node level

show abstract

“…Recent work on citation graphs can be found in [4] where a study is carried out on the citation graph of Computer Science Literature and [31]. In [4], an attempt is made to extract a descriptive summary of the graph through a study of fundamental and well-established properties (degree distribution, giant component size etc.).…”

Section: Related Workmentioning

confidence: 99%