Deep Interest-Shifting Network with Meta-Embeddings for Fresh Item Recommendation

Table 2

The statistics of the datasets.

Datasets	users	items	interactions (M)	auxiliary (M)	warm-start (M)	cold-start
Book-Crossing	0.28 M	0.27 M	1.15	0.75	0.19	48,434
MovieLens	6,040	3,706	1.00	0.76	0.19	30,645
Taobao-Fresh	4.78 M	1.61 M	207.54	203.13	4.09	0.31 M

The first three columns list the number of users, items, and data examples of the whole dataset. The second three columns report the number of data examples of the preprocessed datasets after splitting.