Joint extraction of entities and relations is an important task in information extraction. To tackle this problem, we firstly propose a novel tagging scheme that can convert the joint extraction task to a tagging problem. Then, based on our tagging scheme, we study different end-toend models to extract entities and their relations directly, without identifying entities and relations separately. We conduct experiments on a public dataset produced by distant supervision method and the experimental results show that the tagging based methods are better than most of the existing pipelined and joint learning methods. What's more, the end-to-end model proposed in this paper, achieves the best results on the public dataset.
Hybrid simulations with the global kinetic-magnetohydrodynamic (MHD) code M3D-K have been carried out to investigate the linear stability and nonlinear dynamics of beam-driven fishbone in the experimental advanced superconducting tokamak (EAST) experiment. Linear simulations show that a low frequency fishbone instability is excited at experimental value of beam ion pressure. The mode is mainly driven by low energy beam ions via precessional resonance. The results are consistent with the experimental measurement with respect to mode frequency and mode structure. When the beam ion pressure is increased to exceed a critical value, the low frequency mode transits to a beta-induced Alfvén eigenmode (BAE) with much higher frequency. This BAE is driven by higher energy beam ions. Nonlinear simulations show that the frequency of the low frequency fishbone chirps up and down with corresponding hole-clump structures in phase space, consistent with the Berk–Breizman theory. In addition to the low frequency mode, the high frequency BAE is excited during the nonlinear evolution. For the transient case of beam pressure fraction where the low and high frequency modes are simultaneously excited in the linear phase, only one dominant mode appears in the nonlinear phase with frequency jumps up and down during nonlinear evolution.
a b s t r a c tAn efficient intermittent ultrasonic treatment strategy was developed to improve laccase production from Trametes versicolor mycelia cultures. The optimized strategy consisted of exposing 2-day-old mycelia cultures to 5-min ultrasonic treatments for two times with a 12-h interval at the fixed ultrasonic power and frequency (120 W, 40 kHz). After 5 days of culture, this strategy produced the highest extracellular laccase activity of 588.9 U/L among all treatments tested which was 1.8-fold greater than the control without ultrasound treatment. The ultrasonic treatment resulted in a higher pellet porosity that facilitated the mass transfer of nutrients and metabolites from the pellets to the surrounding liquid. Furthermore, the ultrasonic treatment induced the expression of the laccase gene (lcc), which correlated with a sharp increase in both extracellular and intracellular laccase activity. This is the first study to find positive effects of ultrasound on gene expression in fungal cells. These results provide a basis for understanding the stimulation of metabolite production and process intensification by ultrasonic treatment in filamentous fungal culture.
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.