| Popis: |
The existing biclustering algorithms for finding feature relation based biclusters often depend on assumptions like monotonicity or linearity. Though a few algorithms overcome this problem by using density-based methods, they tend to miss out on many biclusters because they use global criteria for identifying dense regions. The proposed method, RelDenClu, uses the local variations in marginal and joint densities for each pair of features to find the subset of observations, forming the bases of the relation between them. It then finds the set of features connected by a common set of observations, resulting in a bicluster. To show the effectiveness of the proposed methodology, experimentation has been carried out on fourteen simulated datasets. For all the simulated datasets the performance of RelDenClu is compared with that of seven different state-of-the-art algorithms and it is seen to produce better results. Experiments were conducted with the credit card dataset, where binary features were added using RelDenClu. We find that features generated by RelDenClu resulted in improved classification for identifying defaulters. The efficacy of the proposed algorithm is also seen by its use on the COVID-19 dataset for identifying some demographic features that are likely to affect the spread of COVID-19. |
