Personal Recommendation Using a Novel Collaborative Filtering Algorithm in Customer Relationship Management

Table 2

Algorithmic performance for Book-Crossing dataset. The ranking score, precision, recall, intrasimilarity, and hamming distance are corresponding to L = 50,60,70. And the value of parameter 𝛼 is 1.86. Each number presented in this table is obtained by averaging over five runs, each of which has an independently random division of training set and test.

Algorithms	Ranking score	Precision	Recall	Intrasimilarity	Hamming distance
𝐿=50
CF	0.056	0.039	0.181	0.374	0.519
MCF	0.052	0.044	0.192	0.338	0.547
NBI	0.049	0.047	0.204	0.316	0.613
CF-M	0.043	0.051	0.217	0.263	0.764
𝐿=60
CF	0.046	0.037	0.201	0.395	0.511
MCF	0.044	0.042	0.214	0.352	0.536
NBI	0.042	0.045	0.233	0.331	0.604
CF-M	0.037	0.048	0.248	0.278	0.752
𝐿=70
CF	0.042	0.032	0.228	0.421	0.497
MCF	0.040	0.036	0.243	0.375	0.522
NBI	0.034	0.041	0.262	0.349	0.587
CF-M	0.030	0.044	0.296	0.293	0.731