Chapter 9. Advanced Topics
WHAT'S IN THIS CHAPTER?
Using FAST search scalability, for indexing more documents and handling more searches
Reviewing the various types of taxonomies, typical taxonomy workflow, and taxonomy rule sets
Performing entity extraction and the FAST Structural Analysis Toolkits
Tweaking relevancy in FAST
Implementing unsupervised clustering and noun phrase extraction
Reviewing the requirements for Enterprise class Federated Search
Many applications don't need industrial strength search, but when you've got highly specialized requirements or gigabytes of data, you'll be glad to have access to ESP!
Some people assume that "Advanced Search" means jamming a results list chock full of every search widget under the sun; this is often a bad idea! Remember that contemporary users start out having used an Internet search portal like Yahoo! or Google, so anything on the screen that doesn't fit with that mental model has to be easily understood and safe to ignore.
In the early days of search, there was a high-end search company geared to intelligence analysts. Over the years, their flagship search product evolved to include most of the enhancements requested by these advanced knowledge workers. When "normal users" opened up the search application, however, they experienced "shock and awe." There were way too many things on the screen to digest. Other early engines were heavily influenced by librarians, and by very smart groups of "power users." These products often featured quite ...