Embedding-Based Entity Alignment of Cross-Lingual Temporal Knowledge Graphs.

Entity alignment aims to construct a complete knowledge graph (KG) by matching the same entities in multi-source KGs. Existing researches on entity alignment mainly focuses on static multi-relational data in knowledge graphs. However, the relationships or attributes between entities often possess temporal characteristics as well. Neglecting these temporal characteristics can frequently lead to alignment errors. Compared to studying entity alignment in temporal knowledge graphs, there are relatively few efforts on entity alignment in cross-lingual temporal knowledge graphs. Therefore, in this paper, we put forward an entity alignment method for cross-lingual temporal knowledge graphs, namely CTEA. Based on GCN and TransE, CTEA combines entity embeddings, relation embeddings and attribute embeddings to design a joint embedding model, which is more conducive to generating transferable entity embedding. In the meantime, the distance calculation between elements and the similarity calculation of entity pairs are combined to enhance the reliability of cross-lingual entity alignment. Experiments shows that the proposed CTEA model improves Hits@m and MRR by about 0.8∼2.4 percentage points compared with the latest methods.

Cyclic loading changes the taproot's tensile properties and reinforces the soil via the shrub's taproot in semi-arid areas, China.

This study aimed to reveal the soil reinforcement by shrub root systems after repeated stress from external forces, such as high winds and runoff, for extended periods in the wind-hydraulic compound erosion zone. Using the widely distributed Shandong mine area soil and water-conserving plant species, Caragana microphylla, Hippophae rhamnoides, and Artemisia ordosica, cyclic loading tests were conducted on taproots of the three plant species (1-5 mm diameter) via a TY8000 servo-type machine to investigate the taproots' tensile properties response to repeated loading-unloading using simulated high wind pulling and runoff scouring. Our study revealed that the tensile force was positively correlated with the root diameter but the tensile strength was negatively correlated under monotonic and cyclic loading of the three plants' taproots. However, after cyclic loading, the three plant species' taproots significantly enhanced the tensile force and strength more than monotonic loading (P < 0.05). The taproot force-displacement hysteresis curves of the three plant species revealed obvious cyclic characteristics. Structural equation modeling analysis revealed that root diameter and damage method directly affected the taproots' survival rate, reflecting their sustainable soil reinforcement capacity. The damage method significantly influenced the soil reinforcement more than the root diameter. Our findings reveal that the plant species' taproots can adapt more to the external environment and enhance their resistance to erosion after natural low perimeter erosion damage, effectively inducing soil reinforcement. Particularly, the taproots of Caragana microphylla have superior soil-fixing ability and can be used for ecological restoration.

Few-shot link prediction for temporal knowledge graphs based on time-aware translation and attention mechanism.

Few-shot knowledge graph completion (KGC) is an important and common task in real applications, which aims to predict unseen facts when only few samples are available for each relation in the knowledge graph (KG). Previous methods on few-shot KGC mainly focus on static KG, however, many KG in real-world applications are dynamic and develop over time. In this work, we consider few-shot KGC in temporal knowledge graphs (TKGs), where the fact may only hold for a specific timestamp. We propose a Few-Shot Completion model in TKG (TFSC), which compare the input query to the given few-shot references to make predictions. Specifically, in order to enhance the representation of entities in the case of few samples, we use the attention mechanism to model the neighbor entities of the task entity with timestamp information, and generate expressive time-aware entity pair representations through the Transformer encoder. A comprehensive set of experiments is finally carried out to demonstrate the effectiveness a of our proposed model TFSC.

[Automatic Assessment of Facial Nerve Function Based on Infrared Thermal Imaging].

Facial paralysis which is mainly caused by facial nerve dysfunction is a common clinical entity. It seriously devastates a patient's daily life and interpersonal relationships. A method of automatic assessment of facial nerve function is of critical importance for the diagnosis and treatment of facial paralysis. The contralateral asymmetry of facial temperature distribution is one of the newly symptoms of facial paralysis patients which can be captured by infrared thermography. This paper presents a novel framework for objective measurement of facial paralysis based on the automatic analysis of infrared thermal image. Facial infrared thermal image is automatically divided into eight regional areas based on facial temperature distribution specificity and edge detection, the facial temperature distribution features are extracted automatically, including the asymmetry degree of facial temperature distribution, effective thermal area ratio and temperature difference. The automatic classifier is used to assess facial nerve function based on radial basis function neural network (RBFNN). This method comprehensively utilizes the correlation and specificity of the facial temperature distribution，extracts efficiently the facial temperature contralateral asymmetry of facial paralysis in the infrared thermal imaging. In our experiments, 390 infrared thermal images were collected from subjects with unilateral facial paralysis. The results show: the average classification accuracy rate of our proposed method was 94.10%. It has achieved a better classification rate which is above 9.31% than K nearest neighbor (kNN) classifier and 4.87% above Support vector machine (SVM). This experiment results is superior to traditional House-Brackmann facial neural function assessment method. The classification accuracy of facial nerve function with the method is full compliance with the clinical application standard. A complete set of automated techniques for the computerized assessment of thermal images has been developed to assess thermal dysfunction caused by facial paralysis, and the clinical diagnosis and treatment of facial paralysis also will benefit by this method.