The performance of many machine learning algorithms can be substantially improved with a proper discretization scheme. In this paper we describe a theoretically rigorous approach to discretization of continuous attribute values, based on a Bayesian clustering framework. The method produces a probabilistic scoring metric for diierent discretizations, and it can be combined with various types of ...

Discretization is used in click-based graphical passwords so that approximately correct entries can be accepted by the system. We show that the existing discretization scheme of Birget et al.(2006) allows for false accepts and false rejects because the tolerance region is not guaranteed to be centered on the original click-point, causing usability and security concerns. Using empirical data fro...

We present a methodology that enables the use of classification algorithms on regression tasks. We implement this method in system RECLA that transforms a regression problem into a classification one and then uses an existent classification system to solve this new problem. The transformation consists of mapping a continuous variable into an ordinal variable by grouping its values into an appro...

We consider the preservation of weak solution invariants in the time integration of ordinary diier-ential equations (ODEs). Recent research has concentrated on obtaining symplectic discretizations of Hamiltonian systems and schemes that preserve certain rst integrals (i.e. strong invariants). In this article, we examine the connection between constrained systems and ODEs with weak invariants fo...

| Discretization can turn numeric attributes into discrete ones. Feature selection can eliminate some irrelevant and/or redundant attributes. Chi2 is a simple and general algorithm that uses the 2 statistic to discretize numeric attributes repeatedly until some inconsistencies are found in the data. It achieves feature selection via dis-cretization. It can handle mixed attributes, work with mul...

The existence of numeric data and large amounts of records in a database pose a challenging task to explicit concepts extraction from the raw data. This paper introduces a method that reduces data vertically and horizontally, keeps the discriminating power of the original data, and paves the way for extracting concepts. The method is based on discretization (vertical reduction) and feature sele...

A new approach to the conversion of the sampled-data H problem to an equivalent pure discrete-time, lumped-variables counterpart, is presented. The approach is independent of controller causality constraints. In particular, it is applicable to problems, such as preview tracking and fixed-lag smoothing, where existing reduction methods fail. We demonstrate this by solving Open Problem no. 51 fro...

In this paper we transfer the a priori error analysis for the discretization of parabolic optimal control problems on domains allowing for H regularity (i.e. either with smooth boundary or polygonal and convex) to a large class of nonsmooth domains. We show that a combination of two ingredients for the optimal convergence rates with respect to the spatial and the temporal discretization is requ...

Data discretization is an important step in the process of machine learning, since it is easier for classifiers to deal with discrete attributes rather than continuous attributes. Over the years, several methods of performing discretization such as Boolean Reasoning, Equal Frequency Binning, Entropy have been proposed, explored, and implemented. In this article, a simple supervised discretizati...

New methodology for verification of computational methods using unstructured grids is presented. The discretization order properties are studied in computational windows, easily constructed within a collection of grids or a single grid. The windows can be adjusted to isolate the interior discretization, the boundary discretization, or singularities. A major component of the methodology is the d...