A Tree-Based Kernel for Graphs
نویسندگان
چکیده
This paper proposes a new tree-based kernel for graphs. Graphs are decomposed into multisets of ordered Directed Acyclic Graphs (DAGs) and a family of kernels computed by application of tree kernels extended to the DAG domain. We focus our attention on the efficient development of one member of this family. A technique for speeding up the computation is given, as well as theoretical bounds and practical evidence of its feasibility. State of the art results on various benchmark datasets prove the effectiveness of our
منابع مشابه
Shape Similarity Based on a Treelet Kernel with Edition
Several shape similarity measures, based on shape skeletons, are designed in the context of graph kernels. State-of-the-art kernels act on bags of walks, paths or trails which decompose the skeleton graph, and take into account structural noise through edition mechanisms. However, these approaches fail to capture the complexity of junctions inside skeleton graphs due to the linearity of the pat...
متن کاملAn Aligned Subtree Kernel for Weighted Graphs
In this paper, we develop a new entropic matching kernel for weighted graphs by aligning depthbased representations. We demonstrate that this kernel can be seen as an aligned subtree kernel that incorporates explicit subtree correspondences, and thus addresses the drawback of neglecting the relative locations between substructures that arises in the R-convolution kernels. Experiments on standar...
متن کاملA New Heuristic Algorithm for Drawing Binary Trees within Arbitrary Polygons Based on Center of Gravity
Graphs have enormous usage in software engineering, network and electrical engineering. In fact graphs drawing is a geometrically representation of information. Among graphs, trees are concentrated because of their ability in hierarchical extension as well as processing VLSI circuit. Many algorithms have been proposed for drawing binary trees within polygons. However these algorithms generate b...
متن کاملMarginalized Kernels Between Labeled Graphs
A new kernel function between two labeled graphs is presented. Feature vectors are defined as the counts of label paths produced by random walks on graphs. The kernel computation finally boils down to obtaining the stationary state of a discrete-time linear system, thus is efficiently performed by solving simultaneous linear equations. Our kernel is based on an infinite dimensional feature spac...
متن کاملEvaluation of Morphological and Pomological Diversity of 62 Almond Cultivars and Superior Genotypes in Iran
Identification and selection of promising genotypes of fruit tree are primary steps in breeding programs. The economic importance of almond production in the world has stimulated numerous studies related to breeding, quantitative and qualitative traits, the increase of yield and decrease production costs. In this study, morphological and pomological characteristics of 60 cultivar and superior g...
متن کامل