No abstract

Existing knowledge graph (KG) embedding models have primarily focused on static KGs. However, real-world KGs do not remain static, but rather evolve and grow in tandem with the development of KG applications. Consequently, new facts and previously unseen entities and relations continually emerge, necessitating an embedding model that can quickly learn and transfer new knowledge through growth. Motivated by this, we delve into an expanding field of KG embedding in this paper, i.e., lifelong KG embedding. We consider knowledge transfer and retention of the learning on growing snapshots of a KG without having to learn embeddings from scratch. The proposed model includes a masked KG autoencoder for embedding learning and update, with an embedding transfer strategy to inject the learned knowledge into the new entity and relation embeddings, and an embedding regularization method to avoid catastrophic forgetting. To investigate the impacts of different aspects of KG growth, we construct four datasets to evaluate the performance of lifelong KG embedding. Experimental results show that the proposed model outperforms the state-of-the-art inductive and lifelong embedding baselines.

The impact of three arabinoxylan (AX) cultivars (262, 44, 229) on the physicochemical, rheological, and microstructural characteristics of AX hydrogels was explored.The gelation of AX262 was faster and the viscoelastic modulus was higher. During the gelation process, the consumption of ferulic acids was 96%, 87%, and 69% for AX262, AX44, and AX229, respectively, indicating a higher oxidative cross-linking degree of AX262. The decline of adsorption intensity for AX262 hydrogel at 1,644 and 3,423 cm −1 was more pronounced, which might be related to a denser gel network as a result of higher oxidation degree of FA. Meanwhile, AX262 hydrogel was more capable of binding water molecules and exhibited better thermal stability. The difference in the gelation properties of varied AX cultivars lied in the molecular structure.AX262 had a higher ferulic acid content (0.23 μg/mg) and larger molecular weight with 1.17 × 10 6 Da, contributing to an enhanced oxidative cross-linking capacity.Novelty impact statement: Wheat bran arabinoxylans extracted from strong gluten wheat cultivars with higher contents of FA were suitable to fabricate hydrogel system, due to their better oxidative cross-linking capacity. The prepared AX hydrogel had uniform honeycomb structure with enhanced thermal stability and mechanical properties, which could be potentially used as delivery system for bioactive compounds.

No abstract

No abstract