Many statistical queries such as maximum likelihood estimation involve finding the best candidate model given a set of candidate models and a quality esti- mation function. This problem is common in impor- tant applications like land-use classification at multi- ple spatial resolutions from remote sensing raster data. Such a problem is computationally challenging due to the significant computation cost to evaluate the qual- ity estimation function for each candidate model. A recently proposed method of multiscale, multigranular classification has high computational overhead of func- tion evaluation for various candidate models indepen- dently before comparison. In contrast, we propose a context-inclusive approach that controls the computa- tional overhead based on the context, i.e. the value of the quality estimation function for the best candidate model so far. Experimental results using land-use clas- sification at multiple spatial resolutions from satellite imagery show that the proposed approach reduces the computational cost significantly while providing com- parable classification accuracy.