Privacy-Preserving and Scalable Service Recommendation Based on SimHash in a Distributed Cloud Environment

Box 1

Four steps of service recommendation approach .

Step  1 (buliding user indexes offline based on SimHash). For each user , calculate

his/her hash value offline based on SimHash. Then is regarded as the index

for .

Step  2 (finding “probably similar” friends of the target user). According to the same hash

function adopted in Step  1, calculate user index for , that is, . If the Hamming

Distance between and is smaller than 3, then is considered as a

“probably similar” friend of .

Step  3 (finding “really similar” friends of the target user). For a “probably similar” friend

obtained in Step  2, calculate his/her similarity with ; if the similarity is larger than a

threshold , then is a “really similar” friend of .

Step  4 (service recommendation). According to ’s “really similar” friends derived in

Step  3, predict the quality of services never invoked by and recommend the

quality-optimal services to .