PDF Publication Title:
Text from PDF Page: 141
A full comparison of these packages is beyond the scope of this chapter. The only package with the breadth of MatlabBGL is the bioinformatics graph package, which is also based on the Boost graph library. Among the na- tive Matlab packages, only meshpart [Gilbert and Teng, 2002] handles large graphs well. Both MatlabBGL and gaimc are distinguished because they • scaletolargegraphs; • support no-data-copy paths when possible; and • provideasuiteofalgorithms. 6.3 matlabbgl MatlabBGL is the first package we discuss. Its source code lives publicly on LaunchPad, http://launchpad.net/matlab-bgl. As previously mentioned, MatlabBGL is a Matlab package for working with graphs. It uses the Boost graph library to implement the graph algo- rithms efficiently. MatlabBGL is designed to compute on large sparse graphs with hundreds of thousands of nodes. To do so, the library consists of “wrap- pers” for algorithms from the Boost graph library. Each wrapper is a mex function and it is callable directly from Matlab. The goal of the library was to introduce as little new material into Matlab as possible. To facilitate this, MatlabBGL does not introduce a new data structure and uses the Matlab sparse matrix type as the graph type directly. For example, n = 10; A = spdiags([ones(n,1), zeros(n,1), ones(n,1)],[-1 0 1], n, n); cc = clustering_coefficients(A)’ % transpose the output for display comps = components(A)’ % transpose the output for display constructs a 10-node line graph as a Matlab sparse matrix, computes cluster- ing coefficients with the MatlabBGL clustering_coefficients function, and computes the index of a strongly connected component for each vertex. For both of these functions, the output consists of one number per vertex. For clustering_coefficients, it is the clustering coefficient of that vertex, and for components, it is the index of the strong component for that vertex. Examining the output cc = 0000000000 comps = 1111111111 shows that the clustering coefficient for each node is 0, which is expected for the line graph A, and that each node is in the same connected component, which is also expected for a line graph. To describe the library and the remainder of the implementation, we begin by reviewing the Boost graph library. 6.3 ⋅ matlabbgl 121PDF Image | MODELS AND ALGORITHMS FOR PAGERANK SENSITIVITY
PDF Search Title:
MODELS AND ALGORITHMS FOR PAGERANK SENSITIVITYOriginal File Name Searched:
gleich-pagerank-thesis.pdfDIY PDF Search: Google It | Yahoo | Bing
Cruise Ship Reviews | Luxury Resort | Jet | Yacht | and Travel Tech More Info
Cruising Review Topics and Articles More Info
Software based on Filemaker for the travel industry More Info
The Burgenstock Resort: Reviews on CruisingReview website... More Info
Resort Reviews: World Class resorts... More Info
The Riffelalp Resort: Reviews on CruisingReview website... More Info
| CONTACT TEL: 608-238-6001 Email: greg@cruisingreview.com | RSS | AMP |