KeyTEx - An Integrated Prototype for Semi-automatic Metadata Assignment and Network-based Content Retrieval

Benlian, Alexander ; Wiedemann, Florian and Hess, Thomas (December 2004): KeyTEx - An Integrated Prototype for Semi-automatic Metadata Assignment and Network-based Content Retrieval. 10th Americas Conference on Information Systems (AMCIS 2004), New York, USA, August 6-8, 2004. In: AMCIS 2004 Proceedings, 407 pp. 3228-3238

Abstract

In print and online media companies, editors and archivists often have difficulties with assigning relevant metadata to written text in a timely manner. Additionally, they often struggle with finding adequate and useful media content in hierarchically structured archives when searching for background material. In this paper, we describe the potentials of and synergies between text mining and XML topic map technologies for more efficient metadata assignment and media content retrieval processes on the basis of the open source-based Java-prototype KeyTEx1.By illustrating the architecture and relevant functions of the prototype, we show how both types of technology can fit together. Finally, empirical evidence on the economic application of KeyTEx is provided by presenting first findings of a lab experiment. We conclude with assumptions on the relationship between browsing performance and underlying information topology in dependence of the given search setting.

• Item Type: Conference or Workshop Item (Paper)
• Keywords: Text mining; metadata assignment; keyphrase extraction; XML topic maps; information retrieval; semantic networks
• Faculties: Munich School of Management; Munich School of Management > Institute for Digital Management and New Media
• Subjects: 300 Social sciences > 330 Economics
• Language: English