<p>The aim of link prediction is to predict missing links or eliminate spurious links and new links in future network. Among different types of link prediction algorithms, non-negative matrix factorization(NMF)-based methods have become a promising and competitive algorithm with the advantages of dimension reduction and higher prediction accuracy. However, existing NMF-based algorithms have some problems as follows: (1) NMF failure directly captures the neighbor capability of the node; (2) most NMF-based methods do not guarantee that the additional information can effectively improve the performance. To dismiss the above limitations, a novel link prediction model named Pairwisely Constrained Symmetric Non-Negative Matrix Factorization via Degree-related clustering coefficient and Improved closeness centrality (PCSNMF<InlineEquation ID="IEq1"> <InlineMediaObject> <ImageObject Color="BlackWhite" FileRef="10586_2024_4937_Article_IEq1.gif" Format="GIF" Height="6" Rendition="HTML" Resolution="72" Type="Linedraw" Width="14" /> </InlineMediaObject> <EquationSource Format="TEX">\(\_\)</EquationSource> <EquationSource Format="MATHML"><math> <mi>_</mi> </math></EquationSource> </InlineEquation>DI) is proposed in this paper. Specifically, PCSNMF<InlineEquation ID="IEq2"> <InlineMediaObject> <ImageObject Color="BlackWhite" FileRef="10586_2024_4937_Article_IEq1.gif" Format="GIF" Height="6" Rendition="HTML" Resolution="72" Type="Linedraw" Width="14" /> </InlineMediaObject> <EquationSource Format="TEX">\(\_\)</EquationSource> <EquationSource Format="MATHML"><math> <mi>_</mi> </math></EquationSource> </InlineEquation>DI has the following features: (1) the pairwisely constrained graph regularized coupled with degree-related clustering coefficients and improved closeness centrality to maintain both local and global information; and (2) an ablation study is employed to analyse and discuss the necessity of the proposed model for attaching local and global information. In addition, the proof of convergence and the computational complexity for PCSNMF<InlineEquation ID="IEq3"> <InlineMediaObject> <ImageObject Color="BlackWhite" FileRef="10586_2024_4937_Article_IEq1.gif" Format="GIF" Height="6" Rendition="HTML" Resolution="72" Type="Linedraw" Width="14" /> </InlineMediaObject> <EquationSource Format="TEX">\(\_\)</EquationSource> <EquationSource Format="MATHML"><math> <mi>_</mi> </math></EquationSource> </InlineEquation>DI are presented. PCSNMF<InlineEquation ID="IEq4"> <InlineMediaObject> <ImageObject Color="BlackWhite" FileRef="10586_2024_4937_Article_IEq1.gif" Format="GIF" Height="6" Rendition="HTML" Resolution="72" Type="Linedraw" Width="14" /> </InlineMediaObject> <EquationSource Format="TEX">\(\_\)</EquationSource> <EquationSource Format="MATHML"><math> <mi>_</mi> </math></EquationSource> </InlineEquation>DI and baseline methods are evaluated on sixteen real-world networks and the results demonstrate that PCSNMF<InlineEquation ID="IEq5"> <InlineMediaObject> <ImageObject Color="BlackWhite" FileRef="10586_2024_4937_Article_IEq1.gif" Format="GIF" Height="6" Rendition="HTML" Resolution="72" Type="Linedraw" Width="14" /> </InlineMediaObject> <EquationSource Format="TEX">\(\_\)</EquationSource> <EquationSource Format="MATHML"><math> <mi>_</mi> </math></EquationSource> </InlineEquation>DI significantly outperforms state-of-the-art link prediction approaches.</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

Pairwisely constrained symmetric non-negative matrix factorization for link prediction in complex networks using degree-related clustering coefficient and improved closeness centrality

  • Guangfu Chen,
  • Bin Xie,
  • Xiaofei Li,
  • Jizhi Zhao

摘要

The aim of link prediction is to predict missing links or eliminate spurious links and new links in future network. Among different types of link prediction algorithms, non-negative matrix factorization(NMF)-based methods have become a promising and competitive algorithm with the advantages of dimension reduction and higher prediction accuracy. However, existing NMF-based algorithms have some problems as follows: (1) NMF failure directly captures the neighbor capability of the node; (2) most NMF-based methods do not guarantee that the additional information can effectively improve the performance. To dismiss the above limitations, a novel link prediction model named Pairwisely Constrained Symmetric Non-Negative Matrix Factorization via Degree-related clustering coefficient and Improved closeness centrality (PCSNMF \(\_\) _ DI) is proposed in this paper. Specifically, PCSNMF \(\_\) _ DI has the following features: (1) the pairwisely constrained graph regularized coupled with degree-related clustering coefficients and improved closeness centrality to maintain both local and global information; and (2) an ablation study is employed to analyse and discuss the necessity of the proposed model for attaching local and global information. In addition, the proof of convergence and the computational complexity for PCSNMF \(\_\) _ DI are presented. PCSNMF \(\_\) _ DI and baseline methods are evaluated on sixteen real-world networks and the results demonstrate that PCSNMF \(\_\) _ DI significantly outperforms state-of-the-art link prediction approaches.