Webcrawlers are the heart of the search engines of the web. However they are pretty general implementations in order to serve many different but overlaping kinds of searches. A web wide search engine optimized for clinical medical information could find a central role in medical informatics. In my Blog I have discussed the idea of an automatically annotated chart that I first heard articulated by Dr Larry Weed back in 1996.
Consider a search engine optimized for clinical medical data. Actors would include Programs needing document annotation with web sources, direct user queries, and administrative entities for configuration, optimization and medical term cataloguing. Now explore the actor system interactions and the actions which result. This approach, part of the unified process is outlined here with links to detail pages as we drill down on the development. This is in its simplest for an implementation or an applied directed graph so a digression exploring this data structure is here as well.
Consider a clinical search engine based upon establish medical nomenclatures. Optimize this system for domain specific
criteria such as diagnostic probability or clinical relevance. The process could involve,
User (human) query.
Host program call for query by (some) context.
System update or configuration request
Activities to be presented as a flow of activity are in the early conceptual stage of development as the Use Case is being explored. This will be documented here as it proceeds.
As development evolves an API is defined to access data structures and algorithms need for search and optimization followed by context driven result output.