In this projected I migrated a system writen in java to a new architechrure and assessed our algorithm for multidimensional projection in the context of social networks. The placement algorithm can use node's attributes and connectivity pattern to position the network according to the node's similarities. The meaning of similarity depends on the context and is flexible to the user to choose which attributes/edges are used to compute it.
Summary
Publication
Martins, R. M., Andery, G. F., Heberle, H., Paulovich, F. V., De Andrade Lopes, A., Pedrini, H., & Minghim, R. (2012).
Multidimensional projections for visual analysis of social networks.
Journal of Computer Science and Technology, 27(4), 791–810.
Gallery
Data and Results
(A, B, E) Intensities of proteins in each sample; (B, E) after feature selection. (C) Venn diagram and (D) set similarity compare the three ranking approaches.
Sample similarities
Considering all features (A) and selected features (B, C, D).
Sample similarities
Considering all features (A) and selected features (B, C, D).
Sample similarities
Considering all features (A) and selected features (B, C, D).
Sample similarities
Considering all features (A) and selected features (B, C, D).
Sample similarities
Considering all features (A) and selected features (B, C, D).
Sample similarities
Considering all features (A) and selected features (B, C, D).
Sample similarities
Considering all features (A) and selected features (B, C, D).
Sample similarities
Considering all features (A) and selected features (B, C, D).