Supplementing product information by extracting attribute values from title is a crucial task in e-Commerce domain. Previous studies treat each attribute only as an entity type and build one set of NER tags (e.g., BIO) for each of them, leading to a scalability issue which unfits to the large sized attribute system in real world e-Commerce. In this work, we propose a novel approach to support value extraction scaling up to thousands of attributes without losing performance: (1) We propose to regard attribute as a query and adopt only one global set of BIO tags for any attributes to reduce the burden of attribute tag or model explosion; (2) We explicitly model the semantic representations for attribute and title, and develop an attention mechanism to capture the interactive semantic relations in-between to enforce our framework to be attribute comprehensive. We conduct extensive experiments in real-life datasets. The results show that our model not only outperforms existing state-of-the-art N-ER tagging models, but also is robust and generates promising results for up to 8, 906 attributes.
Entity alignment aims to identify equivalent entity pairs from different Knowledge Graphs (KGs), which is essential in integrating multi-source KGs. Recently, with the introduction of GNNs into entity alignment, the architectures of recent models have become more and more complicated. We even find two counter-intuitive phenomena within these methods: (1) The standard linear transformation in GNNs is not working well. (2) Many advanced KG embedding models designed for link prediction task perform poorly in entity alignment. In this paper, we abstract existing entity alignment methods into a unified framework, Shape-Builder & Alignment, which not only successfully explains the above phenomena but also derives two key criteria for an ideal transformation operation. Furthermore, we propose a novel GNNs-based method, Relational Reflection Entity Alignment (RREA). RREA leverages Relational Reflection Transformation to obtain relation specific embeddings for each entity in a more efficient way. The experimental results on real-world datasets show that our model significantly outperforms the state-of-the-art methods, exceeding by 5.8%-10.9% on Hits@1. CCS CONCEPTS • Computing methodologies → Knowledge representation and reasoning; Natural language processing; Supervised learning.
Triplet excitons usually do not emit light under ambient conditions due to the spin-forbidden transition rule, thus they are called dark excitons. Herein, triplet excitons in carbon nanodots (CNDs) are brightened by embedding the CNDs into poly(vinyl alcohol) (PVA) films; flexible multicolor phosphorescence films are thus demonstrated. PVA chains can isolate the CNDs, and excited state electron or energy transfer induced triplet exciton quenching is thus reduced; while the formed hydrogen bonds between the CNDs and PVA can restrict vibration/rotation of the CNDs, thus further protecting the triplet excitons from nonradiative recombination. The lifetimes of the flexible multicolor phosphorescence films can reach 567, 1387, 726, and 311 ms, and the longest-lasting phosphorescence film can be observed by naked eyes for nearly 15 s even after bending 5000 times. The phosphorescence films can be processed into various patterns, and a dynamic optical signature concept has been proposed and demonstrated based on the phosphorescence films.
Seeking the equivalent entities among multi-source Knowledge Graphs (KGs) is the pivotal step to KGs integration, also known as entity alignment (EA). However, most existing EA methods are inefficient and poor in scalability. A recent summary points out that some of them even require several days to deal with a dataset containing 200, 000 nodes (DWY100K). We believe over-complex graph encoder and inefficient negative sampling strategy are the two main reasons. In this paper, we propose a novel KG encoder -Dual Attention Matching Network (Dual-AMN), which not only models both intra-graph and cross-graph information smartly, but also greatly reduces computational complexity. Furthermore, we propose the Normalized Hard Sample Mining Loss to smoothly select hard negative samples with reduced loss shift. The experimental results on widely used public datasets indicate that our method achieves both high accuracy and high efficiency. On DWY100K, the whole running process of our method could be finished in 1, 100 seconds, at least 10× faster than previous work. The performances of our method also outperform previous works across all datasets, where 𝐻𝑖𝑡𝑠@1 and 𝑀𝑅𝑅 have been improved from 6% to 13%. CCS CONCEPTS• Computing methodologies → Knowledge representation and reasoning; Natural language processing; Supervised learning.
Abstract. An analytical model for tidal dynamics has been applied to the Yangtze Estuary for the first time, to describe the tidal propagation in this large and typically branched estuary with three-order branches and four outlets to the sea. This study shows that the analytical model developed for a single-channel estuary can also accurately describe the tidal dynamics in a branched estuary, particularly in the downstream part. Within the same estuary system, the North Branch and the South Branches have a distinct tidal behaviour: the former being amplified demonstrating a marine character and the latter being damped with a riverine character. The satisfactory results for the South Channel and the South Branch using both separate and combined topographies confirm that the branched estuary system functions as an entity. To further test these results, it is suggested to collect more accurate and dense bathymetric and tidal information.
Phosphorescent carbon nanodots (CNDs) have generated enormous interest recently, and the CND phosphorescence is usually located in the visible region, while ultraviolet (UV) phosphorescent CNDs have not been reported thus far. Herein, the UV phosphorescence of CNDs was achieved by decreasing conjugation size and in-situ spatial confinement in a NaCNO crystal. The electron transition from the px to the sp2 orbit of the N atoms within the CNDs can generate one-unit orbital angular momentum, providing a driving force for the triplet excitons population of the CNDs. The confinement caused by the NaCNO crystal reduces the energy dissipation paths of the generated triplet excitons. By further tailoring the size of the CNDs, the phosphorescence wavelength can be tuned to 348 nm, and the room temperature lifetime of the CNDs can reach 15.8 ms. As a demonstration, the UV phosphorescent CNDs were used for inactivating gram-negative and gram-positive bacteria through the emission of their high-energy photons over a long duration, and the resulting antibacterial efficiency reached over 99.9%. This work provides a rational design strategy for UV phosphorescent CNDs and demonstrates their novel antibacterial applications.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
334 Leonard St
Brooklyn, NY 11211
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.