Adding Structure to Unstructured Text
Jonathan I. Maletic, Michael L. Collard · 2005
An overview of the authors’ research program in document engineering is presented. Underlying techniques are being developed for agile parsing of unstructured and semi-structured text to extract metadata. XML technologies are leveraged in novel ways to support complex querying, analysis, and transformation of large text bases. New methods for difference analysis are being developed to support document evolution and maintenance. Additionally, advanced information retrieval methods, namely latent semantic indexing, in conjunction with clustering techniques are used to extract high level features and concepts from large corpora.