machine learning - How to improve accuracy of decision tree in matlab -

i have set of data classify them in matlab using decision tree. divide set 2 parts; 1 training data(85%) , other test data(15%). problem accuracy around %90 , not know how can improve it. appreciate if have idea it.

i guess more important question here what's accuracy given domain: if you're classifying spam 90% might bit low, if you're predicting stock prices 90% high!

if you're doing on known domain set , there previous examples of classification accuracy higher yours, can try several things:

• k-fold cross validation
• ensamble learning
• generalized iterative scaling (gis)
• logistic regression