Mapping metadata to DDC classification structures for searching and browsing

Abstract

In this paper, we introduce a metadata visual interface based on metadata aggregation and automatic classification mapping. We demonstrate that it is possible to aggregate metadata records from multiple unrelated repositories, enhance them through automatic classification, and present them in a unified visual interface. The main features of the interface include dynamic querying using DDC classes as filters, interactive visual views of search results and related DDC classes, and drill-down options for searching and browsing in different levels of details. The interface was tested in a user study of 30 subjects. A comparison was done on three modules of the interface, namely `search interface', `hierarchical interface', and `visual interface.' The results indicate that subjects performed well with all the three interfaces, and they had more positive experience with the hierarchical interface than with the search interface and the visual interface.