Polyadenylation Poly(A) is an essential process during messenger RNA (mRNA) maturation in biological eukaryote systems. Identifying Poly(A) signals (PASs) from the genome level is the key to understanding the mechanism of translation regulation and mRNA metabolism. In this work, we propose a deep dual-dynamic context-aware Poly(A) signal prediction model, called multiscale convolution with self-attention networks (MCANet), to adaptively uncover the spatial–temporal contextual dependence information. Specifically, the model automatically learns and strengthens informative features from the temporalwise and the spatialwise dimension. The identity connectivity performs contextual feature maps of Poly(A) data by direct connections from previous layers to subsequent layers. Then, a fully parametric rectified linear unit (FP-RELU) with dual-dynamic coefficients is devised to make the training of the model easier and enhance the generalization ability. A cross-entropy loss (CL) function is designed to make the model focus on samples that are easy to misclassify. Experiments on different Poly(A) signals demonstrate the superior performance of the proposed MCANet, and an ablation study shows the effectiveness of the network design for the feature learning and prediction of Poly(A) signals.

Sentiment and emotion, which correspond to long-term and short-lived human feelings, are closely linked to each other, leading to the fact that sentiment analysis and emotion recognition are also two interdependent tasks in natural language processing (NLP). One task often leverages the shared knowledge from another task and performs better when solved in a joint learning paradigm. Conversational context dependency, multi-modal interaction, and multi-task correlation are three key factors that contribute to this joint paradigm. However, none of the recent approaches have considered them in a unified framework. To fill this gap, we propose a multi-modal, multi-task interactive graph attention network, termed M3GAT, to simultaneously solve the three problems. At the heart of the model is a proposed interactive conversation graph layer containing three core sub-modules, which are: (1) local-global context connection for modeling both local and global conversational context, (2) cross-modal connection for learning multi-modal complementary and (3) cross-task connection for capturing the correlation across two tasks. Comprehensive experiments on three benchmarking datasets, MELD, MEISD, and MSED, show the effectiveness of M3GAT over state-of-the-art baselines with the margin of 1.88%, 5.37%, and 0.19% for sentiment analysis, and 1.99%, 3.65%, and 0.13% for emotion recognition, respectively. In addition, we also show the superiority of multi-task learning over the single-task framework.

In a big data environment, traditional recommendation methods have limitations such as data sparseness and cold start, etc. In view of the rich semantics, excellent quality, and good structure of knowledge graphs, many researchers have introduced knowledge graphs into the research about recommendation systems, and studied interpretable recommendations based on knowledge graphs. Along this line, this paper proposes a scholar recommendation method based on the high-order propagation of knowledge graph (HoPKG), which analyzes the high-order semantic information in the knowledge graph, and generates richer entity representations to obtain users’ potential interest by distinguishing the importance of different entities. On this basis, a dual aggregation method of high-order propagation is proposed to enable entity information to be propagated more effectively. Through experimental analysis, compared with some baselines, such as Ripplenet, RKGE and CKE, our method has certain advantages in the evaluation indicators AUC and F1.

Discover the deep semantics from the massively structured data in knowledge graph and provide reasonable explanations are a series of important foundational research issues of artificial intelligence. However, the deep semantics hidden between entities in knowledge graph cannot be well expressed. Moreover, considering many predicates express fuzzy relationships, the existing reasoning methods cannot effectively deal with these fuzzy semantics and interpret the corresponding reasoning process. To counter the above problems, in this article, a new interpretable reasoning schema is proposed by introducing fuzzy theory. The presented method focuses on analyzing the fuzzy semantic between related entities in a knowledge graph. By annotating the fuzzy semantic features of adjacency predicates, a novel semantic reasoning model is designed to realize the fuzzy semantic extension over knowledge graph. The evaluation, based on both visualization and query experiments, shows that this proposal has advantages over the initial knowledge graph and can discover more valid semantic information.