Integration of a Methodology for Cluster-Based Retrieval in jColibri

Boosting the Performance of CBR Applications with jCOLIBRI

“…We obtained 15 clusters after applying the SOM clustering algorithm to the case base and we configured the system to work with the 7 most similar clusters (according to the case prototypes). We choose this value because in [10] we show that is a reasonable number of clusters that balances the time gain and precision loss. The results of the clustered CBR application are reported in the Noisy clustering linear row in Table 1 (time 35ms, precision 48,7%).…”

“…jCOLIBRI provides an extension, called Thunder [10], that allows CBR developers to manage case bases organized in clusters and incorporates an in-memory organization model based on Self-Organizing Maps (SOM) [12] as clustering technique. This technique has become one of the most used in Soft-Computing clustering [12] thanks to its capability for discovering hidden patterns and managing uncertainly and partial knowledge.…”

“…Both problematical aspects were successfully solved in a methodology [8] that has been adopted in jCO-LIBRI and is based on a procedure for selecting the most suitable configuration according to the user requirements and the data complexity [2]. This methodology is briefly summarized in Figure 3 (see [10] for further details). This Figure characterizes any case retrieval strategy from a clustered case memory.…”

Successful Case-based Reasoning Applications-2

“…To illustrate the DI-CBR methods in colibri we present an application, described in [66], that serves to classify automatically documents from electronic journals into different categories: laws, history, medicine, ... This system is used by librarians to assign proper bibliographic categories when a new text is included into the catalogue.…”

On Dataset Complexity for Case Base Maintenance

“…F 2 . In [2], we also describe 4 additional measures: one is from the Case-Based Reasoning (CBR) literature [7]; we define the other 3 from ideas presented in the CBR literature [12,16]. Table 2 [9,10], and we have also shown these in the Table. For the purposes of this paper, we select just three of the complexity measures, one from each of the categories.…”