In this paper, we present a kernel trick embedded Gaussian Mixture Model (GMM), called kernel GMM. The basic idea is to embed kernel trick into EM algorithm and deduce a parameter ...
Automated static analysis can identify potential source code anomalies early in the software process that could lead to field failures. However, only a small portion of static ana...
Abstract. Regression models are often required for controlling production processes by predicting parameter values. However, the implicit assumption of standard regression techniqu...
Frank Rosenthal, Peter Benjamin Volk, Martin Hahma...
We describe and analyze a new approach for feature ranking in the presence of categorical features with a large number of possible values. It is shown that popular ranking criteria...
Given a binary classification task, a ranker sorts a set of instances from highest to lowest expectation that the instance is positive. We propose a lexicographic ranker, LexRank,...