There is considerable interest in bridging the terminological gap that exis
ts between the way users prefer to specify their information needs and the
way queries are expressed in terms of keywords or text expressions that occ
ur in documents. One of the approaches proposed for bridging this gap is ba
sed on technologies for expert systems. The central idea of such an approac
h was introduced in the context of a system called Rule Based Information R
etrieval by Computer (RUBRIC). In RUBRIC, user query topics (or concepts) a
re captured in a rule base represented by an AND/OR tree. The evaluation of
AND/OR tree is essentially based on minimum and maximum weights of query t
erms for conjunctions and disjunctions, respectively. The time to generate
the retrieval output of AND/OR tree for a given query topic is exponential
in number of conjunctions in the DNF expression associated with the query t
opic. In this paper, we propose a new approach for computing the retrieval
output. The proposed approach involves preprocessing of the rule base to ge
nerate Minimal Term Sets (MTSs) that speed up the retrieval process. The co
mputational complexity of the on-line query evaluation following the prepro
cessing is polynomial in m. We show that the computation and use of MTSs al
lows a user to choose query topics that best suit their needs and to use re
trieval functions that yield a more refined and controlled retrieval output
than is possible with the AND/OR tree when document terms are binary. We i
ncorporate p-Norm model into the process of evaluating MTSs to handle the c
ase where weights of both documents and query terms are non-binary.