"Book Notes" recommendation System (Recommender systems An Introduction) Seventh Chapter Evaluation recommendation system

Basic idea: The data is divided into training set and test set, training model with training set data, test model with test set data. The Division of Training set and test set can be by the dimension of time, or by the dimension of the crowd. Risk: There may be biases for some methods.
Using historical data to evaluate the data into training set test set and N-fold cross-validation according to Time dimension.

There is also the direct use of human evaluation. However, the cost is larger, not on the scale, in practice is not very useful.

Finish.

"Book Notes" recommendation System (Recommender systems An Introduction) Seventh Chapter Evaluation recommendation system