Timely detection of dynamical complexity changes in natural and man-made systems has deep scientific and practical meanings. We introduce a complexity measure for time series： the base-scale entropy. The definition d...Timely detection of dynamical complexity changes in natural and man-made systems has deep scientific and practical meanings. We introduce a complexity measure for time series： the base-scale entropy. The definition directly applies to arbitrary real-word data. We illustrate our method on a practical speech signal and in a theoretical chaotic system. The results show that the simple and easily calculated measure of base-scale entropy can be effectively used to detect qualitative and quantitative dynamical changes.展开更多
The complexity of heart rate variability (HRV) signal can reflect physiological functions and healthy status of heart system. Detecting complexity of the short-term HRV signal has an important practical meaning. We in...The complexity of heart rate variability (HRV) signal can reflect physiological functions and healthy status of heart system. Detecting complexity of the short-term HRV signal has an important practical meaning. We introduce the base-scale entropy method to analyze the complexity of time series. The advantages of our method are its simplicity, ex-tremely fast calculation for very short data and anti-noise characteristic. For the well-known chaotic dynamical sys-tem―logistic map, it is shown that our complexity be-haves similarly to Lyapunov exponents, and is especially effective in the presence of random Gaussian noise. This paper addresses the use of base-scale entropy method to 3 low-dimensional nonlinear deterministic systems. At last, we apply this idea to short-term HRV signal, and the result shows the method could robustly identify patterns generated from healthy and pathologic states, as well as aging. The base-scale entropy can provide convenience in practically applications.展开更多
方面级情感分析旨在识别文本中针对特定方面的情感倾向,然而现有研究仍面临多重挑战:基于BERT的方面级情感分析研究存在语义过拟合、低层级语义利用不足的问题;自注意力机制存在局部信息丢失的问题;多编码层和多粒度语义的结构存在信息...方面级情感分析旨在识别文本中针对特定方面的情感倾向,然而现有研究仍面临多重挑战:基于BERT的方面级情感分析研究存在语义过拟合、低层级语义利用不足的问题;自注意力机制存在局部信息丢失的问题;多编码层和多粒度语义的结构存在信息冗余问题。为此,提出一种融合BERT编码层的多粒度语义方面级情感分析模型(multi-granular semantic aspect-based sentiment analysis model with fusion of BERT encoding layers,MSBEL)。具体地,引入金字塔注意力机制,利用各个编码层的语义特征,并结合低层编码器以降低过拟合;通过多尺度门控卷积增强模型处理局部信息丢失的能力;使用余弦注意力突出与方面词相关的情感特征,从而减少信息冗余。t-SNE的可视化分析表明,MSBEL的情感表示聚类效果优于BERT。此外,在多个基准数据集上将本文模型与主流模型的性能进行了对比,结果显示:与LCF-BERT相比,本文模型在5个数据集上的F1分别提升了1.53%、3.94%、1.39%、6.68%、5.97%;与SenticGCN相比,本文模型的F1平均提升0.94%,最大提升2.12%;与ABSA-DeBERTa相比,本文模型的F1平均提升1.16%,最大提升4.20%,验证了本文模型在方面级情感分析任务上的有效性和优越性。展开更多
文摘Timely detection of dynamical complexity changes in natural and man-made systems has deep scientific and practical meanings. We introduce a complexity measure for time series： the base-scale entropy. The definition directly applies to arbitrary real-word data. We illustrate our method on a practical speech signal and in a theoretical chaotic system. The results show that the simple and easily calculated measure of base-scale entropy can be effectively used to detect qualitative and quantitative dynamical changes.
文摘The complexity of heart rate variability (HRV) signal can reflect physiological functions and healthy status of heart system. Detecting complexity of the short-term HRV signal has an important practical meaning. We introduce the base-scale entropy method to analyze the complexity of time series. The advantages of our method are its simplicity, ex-tremely fast calculation for very short data and anti-noise characteristic. For the well-known chaotic dynamical sys-tem―logistic map, it is shown that our complexity be-haves similarly to Lyapunov exponents, and is especially effective in the presence of random Gaussian noise. This paper addresses the use of base-scale entropy method to 3 low-dimensional nonlinear deterministic systems. At last, we apply this idea to short-term HRV signal, and the result shows the method could robustly identify patterns generated from healthy and pathologic states, as well as aging. The base-scale entropy can provide convenience in practically applications.
文摘方面级情感分析旨在识别文本中针对特定方面的情感倾向,然而现有研究仍面临多重挑战:基于BERT的方面级情感分析研究存在语义过拟合、低层级语义利用不足的问题;自注意力机制存在局部信息丢失的问题;多编码层和多粒度语义的结构存在信息冗余问题。为此,提出一种融合BERT编码层的多粒度语义方面级情感分析模型(multi-granular semantic aspect-based sentiment analysis model with fusion of BERT encoding layers,MSBEL)。具体地,引入金字塔注意力机制,利用各个编码层的语义特征,并结合低层编码器以降低过拟合;通过多尺度门控卷积增强模型处理局部信息丢失的能力;使用余弦注意力突出与方面词相关的情感特征,从而减少信息冗余。t-SNE的可视化分析表明,MSBEL的情感表示聚类效果优于BERT。此外,在多个基准数据集上将本文模型与主流模型的性能进行了对比,结果显示:与LCF-BERT相比,本文模型在5个数据集上的F1分别提升了1.53%、3.94%、1.39%、6.68%、5.97%;与SenticGCN相比,本文模型的F1平均提升0.94%,最大提升2.12%;与ABSA-DeBERTa相比,本文模型的F1平均提升1.16%,最大提升4.20%,验证了本文模型在方面级情感分析任务上的有效性和优越性。
