FOAF-based clustering of handicraft women using ranked features

This paper builds upon the BWEC1 (Business for Women in Women of Emerging Country) research project to improve the socio-economic situation of handicraft women. In this project our principal task is to build data warehouse schema from handicraft women social network. For that, we follow a semi-supervised clustering-based methodology. In this paper, we propose the adaptation of a semi-supervised hierarchical clustering based on ranking mixed features for the FAOF ontology. This later serves as perfect input data for clustering. The main contribution is to use ontology-based similarity measures that combine numerical and nominal variables along different dimensions (instances, attributes, and relation-ships) and to provide a performable clustering algorithm based on ranking features. The evaluation of the used clustering methods in the context of the project emphasizes it effectiveness to generate valid clusters which can be successfully used for extending the data warehouse schema.