Hierarchical Multifocus Navigation in Text Annotation Data

Abstract

We present a new hierarchical multifocus representation- and interaction-technique for the analysis of text annotations. The comparative analysis of annotation data on multiple, distant passages (focus regions) of a long source text raises several scalability problems. In particular, one soon can be confronted with many nested foci on text ranges with sizes of different orders of magnitude. Our solution suggests to delegate the detailed data representation to other views and to concentrate in the presented overview on the organization of the focus regions. The approach consists of three parts: a collection of strips where the annotations are represented in a condensed form and where sibling- and child-foci are manipulated, a map of the resulting tree of foci for direct access, and a work bench that allows to compare deliberate nodes of the tree on a flat layer. We discuss our approach by comparing it with other state-of-the-art multifocus techniques and describe two use cases that deal with relational annotation and with the analysis of annotations on hierarchical text structures respectively.