Assistance with data mining using R (randomForest, Logistic Regression) from acoustic telemetry data

I need someone who really understands Random Forest, Logistic Regression, and similar datamining algorithms, along with the principles behind these analyses, to help me identify preferred fish habitat. Data are from an acoustic telemetry array. I have already created a number of datasets, including the raw positional data spatially joined to associated attribute data created by interpolation, and modeled diurnal presence/absence data with mean attribute data presented as factors. However, in assessing these data I have encountered some errors and anomalous results (perfect prediction in Random Forest with two trees, for example) that does not make sense to me. I need someone to help me figure out what data mining test(s) would be the most appropriate, write some R code, help me understand what the code is doing, and help me interpret the results once I get them.