How does recenderlab from R count the ratings of each item in ratingMatrix?

Question

I recently started using the R recommenderlab package in my research.
This is the recommenderlab document:

http://cran.r-project.org/web/packages/recommenderlab/vignettes/recommenderlab.pdf

There are some examples in this document, but I have a big question.

• First download the recommenderlab package and Jester5k dataset.

   library("recommenderlab") data(Jester5k) 

• Use the very first 1000 records (users) of Jester5k to study. The algorithm of recommendations is POPULAR.

   r <- Recommender(Jester5k[1:1000], method="POPULAR") 

• Then predict the list of recommendations of the 1001st user. List of 5 best items.

   recom <- predict(r, Jester5k[1001], n=5)<br/> as(recom, "matrix") 

output:

 [1] "j89" "j72" "j47" "j93" "j76"<br/> 

• Then I check the rating of 5 points above.

   rating <- predict(r, Jester5k[1001], type="ratings")<br/> as(rating, "matrix")[,c("j89", "j72", "j47", "j93", "j76")] 

output:

 j89 j72 j47 j93 j76<br/> 2.6476613 2.1273894 0.5867006 1.2997065 1.2956333<br/> 

Why is the list of the top 5 "j89" "j72" "j47" "j93" "j76", but the j47 rating is only 0.5867006.

I do not understand.

How recommenderlab calculate grades of each element in ratingMatrix ?

And how does it display the TopN list?

Answer 1

To get a clearer picture of your problem, I suggest you read the following: “ recommendation: the basis for developing and testing recommendation algorithms ”

Why the list of the top 5 "j89" "j72" "j47" "j93" "j76"

You use the popularity method, which means that you select the list of the top 5, based on the most rated elements (counting the number of saves), and not on the highest predicted rating.

How does recenderlab calculate the ratings of each item in ratingMatrix? And how does he create a TopN list?

Predicted rating, recommanderlab calculates them using conventional distance methods (it is not yet clear if it is pearson or cosine, I did not have the opportunity to check this), then it determines the rating, as suggested by Breeseet al. (1998), the average rating plus a weighted coefficient calculated in the neighborhood, you can consider the entire set of training sessions as the neighborhood of any user, so the predicted ratings for any user on the same element will have the same value.

My best. L