Citation Context

... Broadly relevant feature selection methods can be divided in two groups: feature transformation (such as well known principal component analysis), and unsupervised feature selection (see for example =-=[7,23]-=-). Unfortunately, both groups of methods were developed for a setup that differs from the one that we observe in our dataset. Specifically, previous work considers a finite set of objects, where each ...

Citation Context

...signal-tonoise ratio) value, and that exceed threshold values for skewness, for kurtosis and for the Kendall’s τb cross correlation coefficient. This follows a Feature Selection strategy of filtering =-=[38]-=-. • With 6 injury groups there are 15 pairs of groups. For each pair of groups an SNR for each vector element variable is computed. The difference of the means of each variable over each group to anot...

Citation Context

...tractability of the brute-force, iterative process describesabove, the methodology presented in this paper does not reflect the many alternativesimplementations that may include performance benefits (=-=Somol, Novovičová, & Pudil, 2010-=-).s4.3. Step 2: Build the Predictive ModelsUsing the subset of inputs that are able to minimize squared residuals when predictingsother inputs, the next step is to build a predictive model that produc...