Database querying with personalized vocabulary using data summaries

The increasing size of distributed data sources often leads to large amounts of answers in response to user queries (information overload), and to long response times. Moreover, querying systems do not sufficiently take into account and make use of the user-related information (preferences,context,etc.) that constitute their profile. In this, they are far apart from personalized systems that adapt their behavior or service to the user. Although user-adaptive systems are mostly found on the world wide web, the personalization concept is more general. This paper discusses the possibility and implications of combining the use of summaries with a personalized service, in the querying of a database. The personalization aspect consists in allowing the users to choose their own querying vocabulary, by means of linguistic variables, while the use of summaries increases the efficiency of the querying. Several ways of doing so are identified and discussed. Their advantages and drawbacks are exposed. In some cases, querying algorithms have already been proposed. In other cases, the missing querying algorithms are given.