J. D. Blevins

Learn More
The idea of a perfectly competent but resource limited language user is the basis of many models of sentence comprehension. It is widely assumed that linguistic competence is a) uniform; b) generative; c) autonomous; d) automatic and e) constant. It is also believed that the free expression of these properties is frustrated by limits in the availability of(More)
Three example variants of the k-modes algorithm are compared as tools to illustrate the effects of ties on convergence of any k-modes like algorithm. Two types of ties are discussed as well as their affect on the convergence of the 3 variants. These consequences of resolving these ties are shown to greatly affect speed of convergence and quality of results.(More)
Categorical data can easily have ties. This paper identifies two types of ties and studies their effect on the k-modes methods for categorical data. Three commonly used variants of the k-modes algorithm, each of which handles tie breaking and stopping criterion differently, are compared. It is shown via simple yet subtly constructed examples that the(More)
  • 1