Experimental steps. next up previous
Next: Benchmark user-based system. Up: Experimental Procedure Previous: Experimental Procedure

Experimental steps.

We started our experiments by first dividing the data set into a train and a test portion. Before starting full experimental evaluation of different algorithms we determined the sensitivity of different parameters to different algorithms and from the sensitivity plots we fixed the optimum values of these parameters and used them for the rest of the experiments. To determine the parameter sensitivity, we work only with the train data and further subdivide it into a train and test portion and carried on our experiments on them. For conducted a 10-fold cross validation of our experiments by randomly choosing different train and test sets each time and taking the average of the MAE values.



Badrul M. Sarwar
2001-02-19