Friend recommendation is a fundamental service in both social networks and practical applications, and is influenced by user behaviors such as interactions, interests, and activities. In this study, we first conduct in-depth investigations on factors that affect recommendation results. Next, we design Friend++, a hybrid multi-individual recommendation model that integrates a weighted average method (WAM) into the random walk (RW) framework by seamlessly employing social ties, behavior context, and personal information. In Friend++, the first plus signifies recommending a new friend through network features, while the second plus stands for using node features. To verify our method, we conduct experiments on three social datasets crawled from the Sina microblog system (Weibo). Experimental results show that the proposed method significantly outperforms six baseline methods in terms of recall, precision, F1-measure, and MAP. As a final step, we describe a case study that demonstrates the scalability and universality of our method. Through discussion, we reach a meaningful conclusion: although common interests are more important than user activities in making recommendations, user interactions may be the most important factor in finding the most appropriate potential friends. Keywords multi-individual friend recommendation architecture, behavior context analysis, Intimacy degree, random walk framework, social networks Citation Gong J B, Gao X X, Cheng H, et al. Integrating a weighted-average method into the random walk framework to generate individual friend recommendations.

No abstract

No abstract

In this era of exponential growth in the scale of data, information overload has become an urgent problem, and the use of increasingly flexible sensor cloud systems (SCS) for data collection has become a mainstream trend. Recommendation algorithms can search massive data sets to uncover information that meets the needs of users based on their interests. To improve the accuracy of recommendation scoring, this paper proposes a score prediction algorithm that combines deep learning and matrix factorization. To address the problem of sparse scoring data, our study employs a sensor cloud system to collect data information, preprocesses the collected information, and then uses a deep learning model combined with explicit and implicit feedback to generate recommendations. The proposed algorithm, MF-NeuRec, combines fusion matrix decomposition and the NeuRec model score prediction algorithm. The algorithm employs user-based and item-based NeuRec algorithms to extract the feature vectors of users and items under implicit feedback data. The obtained user and item feature vectors are integrated in a certain ratio through the use of matrix decomposition under the display feedback data. The user and item feature vectors obtained by the algorithm are merged and analyzed to predict how users will rate items. Experiments demonstrate that the algorithm can improve the accuracy of recommendations.

No abstract