R Data Analysis Cookbook, Second Edition - Second Edition

Identifying a suitable model

In order to evaluate various models properly, we need to test them by varying the number of items or movies:

1. Store each model in a list with its name and parameters and then define the number of recommendation items for each user:

> models_to_evaluate <- list( IBCF_cos = list(name = "IBCF", param = list(method = "cosine")), IBCF_cor = list(name = "IBCF", param = list(method = "pearson")), UBCF_cos = list(name = "UBCF", param = list(method = "cosine")), UBCF_cor = list(name = "UBCF", param = list(method = "pearson")), random = list(name = "RANDOM", param=NULL) )> n_recommendations <- c(1, 5, seq(10, 100, 10)) 

2. Run and evaluate the models:

> list_results <- evaluate(x = eval_sets, method = models_to_evaluate, ...