Abstract: This paper proposes a novel approach that combines statistical models and support vector machines. A hybrid scheme which appropriately incorporates the advantages of both the generative and discriminant model paradigms is described and evaluated. Support vector machines (SVMs) are trained to divide the whole speakers' space into small subsets of speakers within a hierarchical tree structure. During testing a speech token is assigned to its corresponding group and evaluation using gaussian mixture models (GMMs) is then processed. Experimental results show that the proposed method can significantly improve the performance of text independent speaker identification task. We report improvements of up to 50% reduction in identification error rate compared to the baseline statistical model.
Abstract: Studies on the distribution of traffic demands have
been proceeding by providing traffic information for reducing
greenhouse gases and reinforcing the road's competitiveness in the
transport section, however, since it is preferentially required the
extensive studies on the driver's behavior changing routes and its
influence factors, this study has been developed a discriminant model
for changing routes considering driving conditions including traffic
conditions of roads and driver's preferences for information media. It
is divided into three groups depending on driving conditions in group
classification with the CART analysis, which is statistically
meaningful. And the extent that driving conditions and preferred
media affect a route change is examined through a discriminant
analysis, and it is developed a discriminant model equation to predict a
route change. As a result of building the discriminant model equation,
it is shown that driving conditions affect a route change much more,
the entire discriminant hit ratio is derived as 64.2%, and this
discriminant equation shows high discriminant ability more than a
certain degree.