A novel information theory method for filter feature selection

In this paper, we propose a novel filter for feature selection. Such filter relies on the estimation of the mutual information between features and classes. We bypass the estimation of the probability density function with the aid of the entropic-graphs approximation of Rényi entropy, and the subsequent approximation of the Shannon one. The complexity of such bypassing process does not depend on the number of dimensions but on the number of patterns/samples, and thus the curse of dimensionality is circumvented. We show that it is then possible to outperform a greedy algorithm based on the maximal relevance and minimal redundancy criterion. We successfully test our method both in the contexts of image classification and microarray data classification.

Factors Associated With Women’s Reporting of Intimate Partner Violence in Spain

In Spain, in 2013, the 20% of women who were murdered by their partner had reported him previously. We analyze the 2011 Spanish-Macrosurvey on Gender Violence to identify and analyze the prevalence of and the principal factors associated with reporting a situation of intimate partner violence (IPV) and the main reasons women cite for not filing such reports, or for subsequently deciding to withdraw their complaint. Overall, 72.8% of women exposed to IPV did not report their aggressor. The most frequent reasons for not reporting were not giving importance to the situation (33.9%), and fear and lack of trust in the reporting process (21.3%). The main reasons for withdrawing the complaint were cessation of the violence (20.0%), and fear and threats (18.2%). The probability of reporting increased among women with young children who were abused, prevalence ratio (95% confidence interval [CI]): 2.14 [1.54, 2.98], and those whose mother was abused, prevalence ratio (95% CI): 2.25 [1.42, 3.57]. Always focusing on the need to protect women who report abuse, it is necessary to promote the availability of and access to legal resources especially among women who use them less: women who do not have children and women who do not have previous family exposure to violence.