Evaluating CBR Similarity Functions for BAM Switching in Networks with Dynamic Traffic Profile.

• DOI: 10.5281/zenodo.1291127
• Publication Date: 2018-06-08
• Authors: Eliseu M. Oliveira; Rafael F. Reale; Joberto S. B. Mar...
• Keywords: Similarity (geometry)

In an increasingly complex scenario for network management, a solution that allows configuration in more autonomous way with less intervention of the manager is expected. This paper presents evaluation similarity functions are necessary context using learning strategy finding solutions. The approach considered based on Case-Based Reasoning (CBR) and applied to where different Bandwidth Allocation Models (BAMs) behaviors used must be eventually switched looking best possible operation. this context, it required identify configure adequate function will process recover similar solutions previously considered. introduces functions, explains relevant aspects which plays role and, finally, proof concept specific adopted. Results show was capable get results from existing use case database. As such, CBR technique has proved potentially satisfactory supporting BAM switching decisions mostly driven by dynamics input traffic profile.