Approximate Retrieval from Databases
Three types of imprecision and/or vagueness:
User does not have precise notion of database contents
Data stored in the database is imprecise
Retrieval request is imprecise
Sources of imprecision in the user's model:
incomplete knowledge of the data model
imprecise info on database schema and/or instance
vagueness of user goals
incomplete knowledge about the interaction tools
How to deal with imprecision in the user's model:
Browsing
Heuristic interpretation of user' query
Example-based techniques
Imprecision in stored data can be dealt with
by applying fuzzy sets theory to provide a linguistic
description of the stored imprecise data.
Fuzzy queries and content-based retrieval allow the user to give imprecise
retrieval requests.