Title: Structured sparsity model with spatial similarity regularisation for semantic feature selection

Authors: Weihua Ou; Wenjun Xiao

Addresses: School of Big Data and Computer Science, Guizhou Normal University, Guiyang, Guizhou Province, 550001, China ' School of Physics and Electronic Science, Guizhou Normal University, Guiyang, Guizhou Province, 550001, China

Abstract: In the prediction of brain activity associated with concrete concepts, the main task is to construct a computational model to reveal the neural basis of the concepts. However, the ordinary regression model cannot select desired semantic features and easily over-fitting. To address these problems, in this paper, we propose a structured sparsity model to automatically choose the relevant semantic features by exploiting the sparsity of responses and the spatial relationships between the voxels. Specifically, we require the number of the non-zero responses to be sparse and the responses that two voxels are nearby in the brain to be similar. The constraints do not only regularise the model fitting but also have an interpretation in terms of brain hemodynamics. The experimental results on predicting brain images show the effectiveness of the proposed approach, as well as improved interpretability.

Keywords: structured sparsity model; spatial relationships; semantic features selection; brain image.

DOI: 10.1504/IJAMC.2017.085941

International Journal of Advanced Media and Communication, 2017 Vol.7 No.2, pp.138 - 149

Received: 30 May 2016
Accepted: 13 Sep 2016
Published online: 18 Aug 2017 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article