Precision The ratio of relevant items selected to the number of items selected. $$P=\frac{N_{rs}}{N_s}$$ 20 66.67
Recall The ratio of relevant items selected to the total number of relevant items available. $$R=\frac{N_{rs}}{N_r}$$ 20 66.67
F1 The combination of precision and recall measures, because it is possible to increase one at the expense of the other. $${F}_1=\frac{2 PR}{P+R}$$ 14 46.67
Accuracy rate Ratio of good recommendations to all recommendations. $$AcR=\frac{N_r}{N}$$ 7 23.33
Mean absolute error The divergence between prediction and actual opinion. $$MAE=\frac{\sum \limits_{i=1}^n\left|{p}_i-{r}_i\right|}{N}$$ 5 16.67
Coverage The ratio of items for which a recommender system can provide recommendations. $$Coverage=\frac{N_s}{N}$$ 4 13.33
Root mean squared error Mean Squared Error (MSE) is a metric that penalizes more major errors than minor errors, but does not offer an intuitive scale. RMSE repositions MSE results along a more intuitive scale. $$RMSE=\sqrt{\frac{\sum \limits_{i=1}^n{\left({p}_i-{r}_i\right)}^2}{N}}$$ 1 3.33
Average rating The average rating from all the users. $$AvR=\frac{\sum \limits_{i=1}^n{r}_i}{\# ratings}$$ 1 3.33