A MINING TECHNIQUE FOR EXTRACTION OF PRESENTATION SCHEMA FROM PRESENTATION DOCUMENTS ACCUMULATED IN LABORATORY

Abstract

The main topic addressed in this paper is how to automatically extract the presentation schema, which represents typical heuristics to be shared by laboratory members for composing presentation structure of presentation documents, from a number of the presentation documents the laboratory members have composed. The key idea is to propose a model of the presentation structure, which represents roles of and sequences among presentation slides included in the presentation documents with metadata. Following this model, the presentation schema is defined as a common presentation structure for the members who have the same methodology and/or domain of research in a research group in the laboratory. This paper accordingly introduces a machine learning technique based on association rule mining for automatically extracting the presentation schema from the repository of
the documents accumulated in the research group. In addition, we report experiments with the comparison of the presentation schemas extracted from three different research groups to investigate the validity of the proposed extracting technique.