worthwhile to render. In this context, by employing a clusteringanalysis with the Gaussian mixture model (GMM), we suggest an empirical framework to classify research collaboration activities with developed indicators that carry on the concept of previous
Possible applications of cluster analysis of bibliographic references as a scientometric method are studied. It is shown that cluster analysis made by means of bibliographic coupling byKessler and co-citation byMarshakova-Small present comparable results. Science maps on immunological topics are made. Particularly for historico-scientific studies it is useful to make clusters in rectangular coordinates taking into account the value of citing the document and the year of its publication. It is observed that at the junction points of sciences there is an almost twofold deceleration of the processes of application and spreading of knowledge. It is stated that the problem of information explosion does not exist on the level of new ideas, the number of which is less than 0.1% of the total volume of the published information flow 40% of which is formed by information noise.
Objective This paper aimed to examine the reliability of co-citation clustering analysis in representing the research history of subject
by comparing the results from co-citation clustering analysis with a review written by authorities.
Methods Firstly, the treatment of traumatic spinal cord injury was chosen as an investigated subject to be retrieved the resource
articles and their references were downloaded from Science Citation Index CD-ROM between 1992 and 2002. Then, the highly cited
papers were arranged chronologically and clustered with the method of co-citation clustering. After mapping the time line
visualization, the history and structure of treatment of spinal cord injury were presented clearly. At last, the results and
the review were compared according the time period, and then the recall and the precision were calculated.
Results The recall was 37.5%, and the precision was 54.5%. The research history of traumatic spinal cord injury treatment analyzed
by co-citation clustering was nearly consistent with authoritative review, although some clusters had shorter period than
which was summarized by professionals.
Conclusion This paper concluded that co-citation clustering analysis was a useful method in representing the research history of subject,
especially for the information researchers, who do not have enough professional knowledge. Its demerit of low recall could
be offset by combination this method with other analytic techniques.
Authors:Chang-Ping Hu, Ji-Ming Hu, Yan Gao, and Yao-Kun Zhang
, scholars usually clusteranalysis, multidimensional scaling analysis and factor analysis to conduct co-citation analysis. Compared to the scholars’ traditional qualitative analysis (including individual induction, interviews and other subjective methods
Authors:Dieter Vanderelst, Sara Speybroeck, and Niko Speybroeck
statistical significant difference between the two classes of diseases in the proportions of missing papers. For each year, a Kruskal–Wallis rank sum test yielded a non-significant result [χ 2 (1) < 1.5, P > 0.05].
Authors:Ji-ping Gao, Kun Ding, Li Teng, and Jie Pang
conduct a literature review. Then we introduce the data set and the method HDCCA with its realization steps. In the next part, we construct the hybrid documents co-citation network, and apply both clusteranalysis and network analysis to uncover the
grouped and considered as single keywords.
Another innovative method was “word clusteranalysis”, combining and analyzing “words in title”, “author keywords”, and “KeyWords Plus”. This was used to discover the research trend, or “research hotspots
Authors:George A. Barnett, Catherine Huh, Youngju Kim, and Han Woo Park
more centrally. Eigenvector centrality determines a journal's overall centrality in the network (Bonacich 1972 ).
Another indicator of the structure of the citation pattern is how journals cluster. Clusteranalysis identifies groups within data
Authors:Jean-Charles Lamirel, Claire Francois, Shadi Al Shehabi, and Martial Hoffmann
The information analysis process includes a cluster analysis or classification step associated with an expert validation of the results. In this paper, we propose new measures of Recall/Precision for estimating the quality of cluster analysis. These measures derive both from the Galois lattice theory and from the Information Retrieval (IR) domain. As opposed to classical measures of inertia, they present the main advantages to be both independent of the classification method and of the difference between the intrinsic dimension of the data and those of the clusters. We present two experiments on the basis of the MultiSOM model, which is an extension of Kohonen's SOM model, as a cluster analysis method. Our first experiment on patent data shows how our measures can be used to compare viewpoint-oriented classification methods, such as MultiSOM, with global cluster analysis method, such as WebSOM. Our second experiment, which takes part in the EICSTES EEC project, is an original Webometrics experiment that combines content and links classification starting from a large non-homogeneous set of web pages. This experiment highlights the fact that break-even points between our different measures of Recall/Precision can be used to determine an optimal number of clusters for web data classification. The content of the clusters obtained when using different break-even points are compared for determining the quality of the resulting maps.