Metric for information retrival where positions are discounted logarithmically.
nDCG(recommendedIDX, testSetIDX)indices of the recommended items. Object of class numeric.
indices of the items in the test set. Object of class numeric
nDCG is computed as the ratio between Discounted Cumulative Gain(DCG) and idealized Discounted Cumulative Gain(IDCG): $$DGC_{pos} = rel_1 + \sum_{i=2}^{pos} \frac{rel_i}{\log_2i}$$ $$IDGC_{pos} = rel_1 + \sum_{i=2}^{|h|-1} \frac{rel_i}{\log_2i}$$ $$nDCG_{pos} = \frac{DCG}{IDCG}$$
Asela Gunawardana, Guy Shani, Evaluating Recommender Systems.