Search

User representation learning is crucial for capturing different user preferences, but it is also critical challenging because user intentions are latent and dispersed in complex and different patterns of user-generated data, and thus cannot be measured directly. Text-based data models can learn user representations by mining latent semantics, which is beneficial to enhancing the semantic function of user representations. However, these technologies only extract common features in historical records and cannot represent changes in user intentions. However, sequential feature can express the user's interests and intentions that change time by time. But the sequential recommendation results based on the user representation of the item lack the interpretability of preference factors. To address these issues, we propose in this paper a novel model with Dual-Layer User Representation, named DLUR, where the user's intention is learned based on two different layer representations. Specifically, the latent semantic layer adds an interactive layer based on Transformer to extract keywords and key sentences in the text and serve as a basis for interpretation. The sequence layer uses the Transformer model to encode the user's preference intention to clarify changes in the user's intention. Therefore, this dual-layer user mode is more comprehensive than a single text mode or sequence mode and can effectually improve the performance of recommendations. Our extensive experiments on five benchmark datasets demonstrate DLUR's performance over state-of-the-art recommendation models. In addition, DLUR's ability to explain recommendation results is also demonstrated through some specific cases. [ABSTRACT FROM AUTHOR]

The influence maximization problem is defined by identifying the seed set that has the most influence on other users in the network, which when selected, the cascading process reaches a large number of users. We use a greedy algorithm and an epsilon-greedy algorithm from the MAB models in this work, unlike prior works that used the MAB models to quantify the unknown propagation probability in the diffusion models. In this paper, we did not also make any assumption regarding the diffusion models and tries to learn to identify the most influential users based on designed reward function “hybrid edge strength-similarity” using global centrality measures and by trying to find a tradeoff between exploitation and exploration strategies. The new proposed reward function initializes the MAB algorithms using global characteristics that quantify the strength of each arm (edge). The proposed reward will feed algorithms from MAB models uses hybridization of edge betweenness centrality and Jaccard similarity measures with some level of participation of each measure. Then, three algorithms are proposed for the extraction of relevant influencers, namely: SRI-CGSS FEXPL-GREEDY) algorithm which almost exploiting the best arm; the SRI_CGSS FEXPR-GREEDY which is almost exploring; and the SRI-MAB ε -GREEDY algorithm that alternate between exploring and exploiting the best arms. We conduct extensive experiments on a large-scale graph in terms of influence spread, efficiency performance in terms of running time and space complexity, and how the reward parameters impact cumulative regret.

The genetic causes in most of patients with oocyte maturation arrest remain largely unknown. In this study, we identified a homozygous missense mutation (c.895T>C; p.C299R) in TBPL2 (TATA box binding protein like 2) in two infertile sisters with oocyte maturation arrest and degeneration from a consanguineous family by whole‐exome sequencing. The TBPL2 mutation is rare and pathogenic, and impaired the transcription initiation function of the protein. Our results showed that TBPL2 mutation might be associated with female infertility due to oocyte maturation arrest and degeneration. [ABSTRACT FROM AUTHOR]

Background: Empty follicle syndrome (EFS) is a rare but severe condition in which no oocyte is recovered in female patients undergoing in vitro fertilization (IVF) after sufficient ovarian response to hormonal trigger. Accumulating evidence highlights the genetic basis of EFS occurrence. Methods: In this study, we report a patient with primary infertility showing the characteristics of EFS from a consanguineous family. Under the treatment of assisted reproductive technique (ART), no oocyte was retrieved following the aspiration of mature follicles. Through whole‐exome sequencing (WES), we discovered a novel recessively transmitted mutation in ZP1 (c.769 C>T, p. Q257*). Results: In vitro Co‐immunoprecipitation assays showed that mutant ZP1 protein failed to interact with either ZP2 or ZP3, which explains the degenerated oocytes in the patient with EFS. Conclusion: Together, our data further expand the spectrum of ZP1 mutations that are associated with human EFS and thus provide novel insight into the diagnosis of EFS patients. [ABSTRACT FROM AUTHOR]

In large-scale applications, 5G networks are faced with various challenges such as large-scale mobile terminals, dynamic topology and resource management. To solve these problems, we introduced the edge computing, peer entity perception, network load balancing and calculation offload into 5G networks, and proposed a real-time and efficient edge computing offload mechanism. On the one hand, the mechanism, through organic distribution and deep fusion of mobile terminals and cloud computing servers, rationally allocate storage, computing and network services, could achieve a feasible division of labor and collaboration between the quality of the network service and the quality of the user experience. On the other hand, the mechanism can satisfy the needs of massive mobile terminals and peer entities to access 5G networks by introducing peer to peer entities. In addition, the mechanism removes the mismatch features and isomerism between the mobile terminal of the limited 5G network resources and computing power and the high-performance server in real time through mobile peer sensing. Thus, a balance between user experience requirements and network service quality is sought. Simulation experiments and mathematical analysis results demonstrate the advantages of the proposed algorithm in terms of real-time performance, resource management level and computational efficiency. [ABSTRACT FROM AUTHOR]