Rule-extraction from trained neural networks has previously been used to generate propositional rule-sets. While this is suucient for many applications where large rule sets can be eeectively used, it is clearly desirable to provide a more general explanation capability. In this paper we describe an approach to representing a network as a PROLOG logic program, where the activation values of hid...

The capacity for abstraction is vital for adaptive behavior. Based on behavioral data and event-related potentials (ERPs), the present study investigates the brain's ability to encode abstract auditory rules with a dynamic approach in which rules constantly emerge and vanish. In successive conditions, abstract rules are task-irrelevant and task-relevant. Results show that as few as two consecut...

In a dependent type theory satisfying the propositions as types correspondence together with the proofs-as-programs paradigm, the validity of the unique choice rule or even more of the choice rule says that the extraction of a computable witness from an existential statement under hypothesis can be performed within the same theory. Here we show that the unique choice rule, and hence the choice ...

Obtaining and storing large amounts of data have become easier with the rapidly developing information technologies (IT). However, generated collected, which are irrelevant in themselves, useful only when they analyzed for a specific reason. Data mining may transform raw into information. In present study, classification analysis denim fabric quality characteristics according to production para...

In recent years, support vector machines (SVMs) have shown good performance in a number of application areas, including text classification. However, the success of SVMs comes at a cost – an inability to explain the process by which a learning result was reached and why a decision is being made. Rule-extraction from SVMs is important for the acceptance of this machine learning technology, espec...