Category: String
Organization: EUUG
Abstract: When designing interactive information processing
        applications there are two important properties of
        the data files that must be considered. The first of
        these is the size of the file. The second is speed of
        access. An investigation was conducted into the most
        appropriate structures for the storage of
        dictionaries. Sequential, binary and hashing
        techniques proved to be inferior to tree-based
        methods which permitted data compression. Tree
        structures for storing a dictionary file were used in
        writing an interactive spelling corrector because:
        tree-based storage methods were the only method to
        introduce data compression; access times were found
        to be better in comparison with the sequential search
        and comparable with binary chop or hashing; and it
        was discovered that there existed inherent spelling
        correction within the tree
Bibtype: InProceedings
Keywords: data structures; file organisation; spelling aids;
        trees (mathematics); Unix; interactive UNIX spelling
        corrector; interactive information processing
        applications; data files; storage; dictionaries;
        tree-based methods; data compression; access times;
        binary chop; hashing
Booktitle: Proceedings of the Spring 1989 EUUG Conference
Author: P. M. Sleat
        S. K. Das
Title: An interactive UNIX spelling corrector
Year: 1989