Loading...

Isograph: Neighbourhood graph construction based on geodesic distance for semi-supervised learning

Ghazvininejad, M ; Sharif University of Technology | 2011

866 Viewed
  1. Type of Document: Article
  2. DOI: 10.1109/ICDM.2011.83
  3. Publisher: 2011
  4. Abstract:
  5. Semi-supervised learning based on manifolds has been the focus of extensive research in recent years. Convenient neighbourhood graph construction is a key component of a successful semi-supervised classification method. Previous graph construction methods fail when there are pairs of data points that have small Euclidean distance, but are far apart over the manifold. To overcome this problem, we start with an arbitrary neighbourhood graph and iteratively update the edge weights by using the estimates of the geodesic distances between points. Moreover, we provide theoretical bounds on the values of estimated geodesic distances. Experimental results on real-world data show significant improvement compared to the previous graph construction methods
  6. Keywords:
  7. Graph construction ; Data points ; Edge weights ; Euclidean distance ; Graph-construction method ; Key component ; Manifold ; Neighbourhood graphs ; Real world data ; Semi-supervised classification method ; Semi-supervised learning ; Theoretical bounds ; Data mining ; Geodesy ; Supervised learning
  8. Source: Proceedings - IEEE International Conference on Data Mining, ICDM, 11 December 2011 through 14 December 2011 ; December , 2011 , Pages 191-200 ; 15504786 (ISSN) ; 9780769544083 (ISBN)
  9. URL: http://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=6137223