Next: Documents
Up: Hybrid-Search and Storage of
Previous: Lore
Before we can continue with a formal discussion of the implementation
of hybrid-search in Haystack, it would be advantageous to understand
exactly why hybrid-search is an appropriate method for searching
through documents. This chapter will create a model for documents
based on state. By analyzing the structure of this model, and the
type of queries a user may invoke against it, a data model for
Haystack will be developed. We will then show how to index this data
structure into a hybrid-search system.
Copyright 1998, Eytan Adar (eytan@alum.mit.edu)