An Analysis of Current Trends in CBR Research Using Multiview Clustering

In this report we review the research themes covered in these papers and identify the topics that are active at the moment. The main mechanism for this analysis is a clustering of the research papers based on both cocitation links and text similarity. It is interesting to note that the core set of papers has attracted citations from almost three thousand papers outside the conference collection so it is clear that the CBR conferences are a subpart of a much larger whole. It is remarkable that the research themes revealed by this analysis do not map directly to the subtopics of CBR that might appear in a textbook. Instead they reflect the applications-oriented focus of CBR research, and cover the promising application areas and research challenges that are faced. Each year since 1993 there has been an international or European conference on CBR. Up to 2007, this conference series produced 672 papers in all. In this report we examine the research themes evident in these papers and identify the most active research topics in CBR. At the 2008 conference we presented an analysis of the research themes in CBR, based on an analysis of the cocitation links in the research literature (Greene et al. 2008). That analysis was based on the core set of 672 papers from the CBR conferences with cocitation data coming from a set of 3461 papers that cite these papers (details on how cocitation links are determined are given later in the article). While cocitation analysis has been proven to be very effective at uncovering relational structure in the research literature (White and Griffith 1981), it has the shortcoming that recent papers will have few cocitation links as papers citing pairs of papers in the core set (that is, the source of cocitation links) have not yet appeared. This issue is evident in the plot of citation counts shown in figure 1 and ultimately makes it impossible to recognize the influence of more recent papers.