Weize Chen, Xu Han, Yankai Lin, Hexu Zhao, Zhiyuan Liu, Peng Li, Maosong Sun, Jie Zhou. Proceedings of the 60th Annual Meeting Association for Computational Linguistics (Volume 1: Long Papers). 2022.

Monitoring the condition of a dc-link capacitor is important for reliability-critical voltage source inverter (VSI) systems. With ever increasing application metalized poly propylene film (MPPF) capacitor, new challenge monitoring (CM) presented due to very small equivalent series resistance and capacitance end-of-life criterion (typically 2%-5%) an MPPF capacitor. To improve accuracy estimation, novel quasi-online CM method proposed in this paper, which introduces LC resonance intentionally...

Joint representation learning of text and knowledge within a unified semantic space enables us to perform graph completion more accurately. In this work, we propose novel framework embed words, entities relations into the same continuous vector space. model, both entity relation embeddings are learned by taking plain consideration. experiments, evaluate joint model on three tasks including prediction, prediction classification from text. The experiment results show that our can significantly...

We present a statistical framework to extract information-rich citation sentences that summarise the main contributions of scientific paper. In first stage, we automatically discover salient keywords from paper’s summary, characterise its contributions. asecond exploitingthe results identify best capture Experimental show our approach using methods rooted in quantitative statistics and information theory outperforms current state-of-the-art systems paper summarisation.

The increasingly prevalent use of chatbots provides an innovative way to conduct conversation interviews. However, the lack comprehensive design guidance and robust evaluation methodologies challenge designers in developing interview with good elicitation ethics. This proposal presents our progress on investigating chatbots' performances eliciting high-quality response without breaking any ethical rules, based which automatic framework will be developed. We also present plan propose...

A scientific domain consists of subfields that can be further refined into specialisations. Specialisations emerge, evolve and consolidate, as reflected in particular literature development, along a contents-based dimension where important problems are stated addressed, communal researchers collaborate compete to solve those problems. We propose generic framework aims at effectively identifying characterising the main specialisations by leveraging both paper contents citation links. More...

Abstract Older adults may be excluded from using digital health technologies due to limited eHealth literacy. Research is much needed decrease disparities in literacy and increase the inclusiveness of such technologies. Integrating preferences expertise older key age‐appropriate design tutorials. This study explores how participatory (PD) techniques can adapted include an tutorial. We worked with 9 (aged 64 82) as co‐designers conducted PD sessions over 11 weeks a senior center's computer...

Against the background of rapid development artificial intelligence (AI), this article reviews process education and teaching models, refines measures for digital transformation in education, analyzes current situation nuclear engineering education. It concludes that will inevitably face transformation. Consequently, a preliminary attempt was made to build twin platform (i.e. reactor platform). The introduction courses has greatly improved students' learning enthusiasm practical abilities,...

In this paper, we present a novel self-supervised framework for Sentiment Cue Extraction (SCE) aimed at enhancing the interpretability of text sentiment analysis models. Our approach leverages learning to identify and highlight key textual elements that significantly influence classification decisions. Central our is development an innovative Mask Sequence Interpretation Score (MSIS), bespoke metric designed assess relevance coherence identified cues within binary tasks. By employing Monte...

From an evolutionary perspective, a body of research is evolving ecosystem, consisting topics subjected to form natural selection as come into existence, and thrive more or less over variable period time. Identifying the establishment given topic in scientific domain, terms its momentum at time inquiry, can provide useful insights where this heading, facilitate effective literature research. Here we propose identify three forms topics, emerging from comparison between two different...

At present, the publishing organization's publication selection planning method lags behind, making China's industry economic benefits greatly reduced. This paper presents a new affective analysis method-Affective Recommendation index (ARIA) based on of fine-grained analysis, which combines conditional random field model (CRF) and weighting to analyze evaluation information publication, Targeted object its attributes reader's emotional inclination, taking full account emotion-free words but...

This paper presents a centralized coordinated voltage control method for active distribution network to solve off-limit problem of after incorporation distributed generation (DG). The proposed consists two parts, it primal-dual interior point method-based regulation schemes DG reactive powers and capacitors with on-load tap changer (OLTC) controlling which utilizes system’s maximum minimum voltages, improve the qualified rate reduce operation numbers OLTC. coordination has considered cost...

In this paper, we develop a low than character feature embedding called radical embedding, and apply it on LSTM model for sentence segmentation of pre modern Chinese texts. The datasets includes over 150 classical books from 3 different dynasties contains literary styles. CRF is state art method the sequence labeling problem. Our new adds component which leads to improved performances. Experimental results based aforementioned demonstrates better accuracy earlier methods segmentation,...

Film translation has experienced a long history in China. In 1949, the film Private Aleksandr Matrosov was first to be imported and translated into Chinese from former Soviet Union. This continues contemporary era where China become one of biggest most prosperous markets globally; hundreds international movies are compete at office box for market shares. The functions, aesthetic standards, values translation, including dubbing subtitles, meanwhile, have changed along with different...