The significant growth of consumer credit has resulted in a wide range of statistical and non-statistical methods for classifying applicants in ‘good’ and ‘bad’ risk categories. Traditionally, (logistic) regression used to be one of the most popular methods for this task, but recently some newer techniques like neural networks and support vector machines have shown excellent classification perf...

This article addresses the problem of the induction of comprehensible hypotheses on complex domains. It proposes a hypothesis language (a type of decision graph) and an inductive algorithm (PASTEUR) which naturally induces structured rules. The advantages over other hypothesis languages in terms of comprehensibility are reviewed as well as the possibility given to represent explicitly exception...

Towards Comprehensible and Effective Permission Systems

The article introduces quantile deviation l as a new sensitivity measure based on the difference between superquantile and subquantile. New global indices square of are presented. proposed compared with quantile-oriented subordinated to contrasts classical Sobol indices. comparison is performed in case study using non-linear mathematical function, output which represents elastic resistance slen...

Large databases are routinely being collected in government, science, business and medicine. A variety of techniques from statistics, signal processing, pattern recognition, machine learning, and neural networks have been proposed to help people understand the data by discovering predictive models of the data. However, research in data mining has not paid attention to the factors that make lear...

Supervised classiication is one of the most active areas of machine learning research. Most work has focused on classiication in static domains, where an instantaneous snapshot of attributes is meaningful. In many domains, attributes are not static; in fact, it is the way they vary temporally that can make classiication possible. Examples of such domains include speech recognition, gesture reco...

We propose a method for generating comprehensible explanations in description logic. Such explanations could be of potential use for e.g. engineers, doctors, and users of the semantic web. Users commonly need to understand why a logical statement follows from a set of hypotheses. Then, automatically generated explanations that are easily understandable could be of help. A proof system for descr...