Scholarlink

<p>One of the challenges in studying time series signals is analyzing functional connectivity between them. This paper presents a novel approach to functional connectivity method among time series, incorporating their intrinsic geometric properties. We used the geodesic distance between time series to construct a geometric functional connectivity analysis (GFCA) method. The new proposed geometric functional connectivity analysis method has been contained three categories. The first category involves Gaussian geodesic distance among time series, the second encompasses the construction of a geometric weighted cross visibility graph, and the third pertains to the introduction of geometric recurrent connectivity measurements. The feature sets extracted based on the geometric functional connectivity analysis have been applied to evaluate the method’s effectiveness in driver distraction detection. The results demonstrate that combining basic functional connectivity feature sets with the newly proposed geometric functional connectivity feature sets statistically significant (p-value<InlineEquation ID="IEq1"> <InlineMediaObject> <ImageObject Color="BlackWhite" FileRef="11760_2025_4692_Article_IEq1.gif" Format="GIF" Height="13" Rendition="HTML" Resolution="72" Type="Linedraw" Width="48" /> </InlineMediaObject> <EquationSource Format="TEX">\(&lt;0.05\)</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mo>&lt;</mo> <mn>0.05</mn> </mrow> </math></EquationSource> </InlineEquation>) improves classification performance.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A Geometric Based Functional Connectivity Analysis for Spatiotemporal Time Series Modelling

  • 期刊论文

  • 发表日期：

    2025年9月11日

查看全文
  • Hajar Ghahremani-Gol,
  • Amin Mohammadian

摘要

One of the challenges in studying time series signals is analyzing functional connectivity between them. This paper presents a novel approach to functional connectivity method among time series, incorporating their intrinsic geometric properties. We used the geodesic distance between time series to construct a geometric functional connectivity analysis (GFCA) method. The new proposed geometric functional connectivity analysis method has been contained three categories. The first category involves Gaussian geodesic distance among time series, the second encompasses the construction of a geometric weighted cross visibility graph, and the third pertains to the introduction of geometric recurrent connectivity measurements. The feature sets extracted based on the geometric functional connectivity analysis have been applied to evaluate the method’s effectiveness in driver distraction detection. The results demonstrate that combining basic functional connectivity feature sets with the newly proposed geometric functional connectivity feature sets statistically significant (p-value \(<0.05\) < 0.05 ) improves classification performance.