R语言 协同过滤调参

协同过滤推荐算法调整参数（Neighborhood Size)时一直报错，请问这是为什么呀？

user_nn <- list(
  "10 NN" = list(name="UBCF", param=list(normalize = "Z-score",
                                         method="Cosine",
                                         nn=10)),
  "20 NN" = list(name="UBCF", param=list(normalize = "Z-score",
                                         method="Cosine",
                                         nn=20)),
  "30 NN" = list(name="UBCF", param=list(normalize = "Z-score",
                                         method="Cosine",
                                         nn=30)),
  "40 NN" = list(name="UBCF", param=list(normalize = "Z-score",
                                         method="Cosine",
                                         nn=40)),
  "50 NN" = list(name="UBCF", param=list(normalize = "Z-score",
                                         method="Cosine",
                                         nn=50)),
  "60 NN" = list(name="UBCF", param=list(normalize = "Z-score",
                                         method="Cosine",
                                         nn=60))
)
# Run the algorithm and predict the next n items for comparison purposes
recs <- c(10,20,30,40,50)
user_nn_results <- evaluate(record, user_nn, n = recs, progress = FALSE)

img

假定你使用的是recommenderlab这个Libary，你的输入有一个矩阵表示成稀疏矩阵了，需要转成普通矩阵。

record <- as.matrix(record)

可以参考这两个英文解答


https://stackoverflow.com/questions/35437411/error-in-predict-glmnet-function-not-yet-implemented-method