Partitioned Feature-based Classifier model with Expertise Table
Dong-Chul Park · 2010
An advanced form of the Partitioned Feature-based Classifier (PFC) is proposed in this paper. As is the case with the PFC, the proposed classifier model, called Partitioned Feature-based Classifier with Expertise Table (PFC-ET), does not use the entire feature vectors extracted from the original data in a concatenated form to classify each datum, but rather uses groups of features related to each feature vector separately. The proposed PFC-ET improves the contribution rate used in the PFC by introducing a confusion table, called an Expertise Table, for each local classifier that uses a specific feature vector group. The confusion table for each local classifier contains accuracy information of each local classifier on each class of data. The proposed PFC-ET algorithm is applied to the problem of music genre classification on a set of music data. The results demonstrate that the proposed PFC-ET model outperforms the original PFC model by 7.22% - 23.6% on average in terms of classification accuracy depending on the grouping algorithms used for local classifiers and the number of clusters.