Microservice‐driven privacy‐aware cross‐platform social relationship prediction based on sequential information

Abstract

Currently, the accurate prediction of social relationships can effectively reduce the decision‐making burden of users in various service platforms. However, in the big data environment, the users' data information used for the relationship prediction is highly fragmented distribution, so it is a non‐trivial challenge to integrate the users' sequence data information from different platforms while preventing sensitive information leakage. To this end, based on the microservice environment, we devise a cross‐platform social relationship prediction approach (CPSRP) to address the above problems. Briefly, the improved Simhash method aggregates similar users into the common bucket. Then the embedding technique converts the users' sparse data information into the low‐dimensional dense continuous feature vectors; the redefined Gated Recurrent Unit (r‐GRU) network and the Multilayer Perceptron (MLP) network are employed to extract the overall temporal sequence features of users. The relationship prediction is finally executed according to the users' sequential features. Extensive experiments are conducted on Epinions, and the experimental results further prove the benefits of our proposal in terms of relationship prediction while protecting users' sensitive information.

FSRPCL: Privacy-preserve federated social relationship prediction with contrastive learning

“…Furthermore, these two methods utilized the Simhash technique and the Locality-Sensitive Hashing technology to find a similar set of users, respectively. In addition, Tang et al [1] proved that there was a strong internal correlation between social relationship prediction and users' similarity through the analysis of users' historical behavior information, thus indirectly showing that the homogeneity theory in sociology explained the reasons for the establishment of social relationships. Although the above methods have achieved certain achievements in social relationship prediction, all of them considered the users' historical behavior information as static information.…”

“…Although our cross-platform social relationship prediction scheme ensures the performance of social relationship prediction while protecting users' private information, it still suffers from the following weaknesses: (1) We have not taken into account the influence of the information about the network structure in social networks on the scenarios of crossplatform social relationship prediction. (2) We intend to explore cross-platform social relationship prediction schemes using horizontal federated learning and/or federated transfer learning.…”

“…on various consumer platforms such as Tmall, Jingdong, and Taobao. According to existing homogeneous sociological theories [1] , it is anticipated that users who share similar historical behavioral attributes, particularly in terms of consumption behavior, are expected to form new social relationships. Consequently, the proliferation of multiple platforms on various intelligent devices has contributed to the intricacy and unpredictability of online social networks.…”

“…Additionally, existing social relationship prediction methods [1,19] typically assume that the users' historical data remains static and overlook the impact of interdependencies among items in the users' behavioral sequence on the users' characteristics. The comprehensive exploration and analysis of the sequence characteristics of the users' behavior is also hindered by this limitation.…”