Inverse Entailment in Nonmonotonic Logic Programs

We present a heuristic based algorithm to induce nonmonotonic logic programs that will explain the behavior of XGBoost trained classifiers. We use the technique based on the LIME approach to locally select the most important features contributing to the classification decision. Then, in order to explain the model’s global behavior, we propose the LIME-FOLD algorithm —a heuristic-based inductive logic programming (ILP) algorithm capable of learning nonmonotonic logic programs—that we apply to a transformed dataset produced by LIME. Our proposed approach is agnostic to the choice of the ILP algorithm. Our experiments with UCI standard benchmarks suggest a significant improvement in terms of classification evaluation metrics. Meanwhile, the number of induced rules dramatically decreases compared to ALEPH, a state-of-the-art ILP system.

Download Full-text

Relevance-Driven Evaluation of Modular Nonmonotonic Logic Programs

Logic Programming and Nonmonotonic Reasoning - Lecture Notes in Computer Science ◽

10.1007/978-3-642-04238-6_10 ◽

2009 ◽

pp. 87-100 ◽

Cited By ~ 1

Author(s):

Minh Dao-Tran ◽

Thomas Eiter ◽

Michael Fink ◽

Thomas Krennwallner

Keyword(s):

Nonmonotonic Logic ◽

Logic Programs

Download Full-text

HOW TO COMBINE NONMONOTONIC LOGIC AND RAPID PROTOTYPING TO HELP MAINTAIN SOFTWARE

International Journal of Software Engineering and Knowledge Engineering ◽

10.1142/s021819409500006x ◽

1995 ◽

Vol 05 (01) ◽

pp. 89-118 ◽

Cited By ~ 10

Author(s):

LUQI ◽

DANIEL E. COOKE

Keyword(s):

Rapid Prototyping ◽

Logic Programming ◽

Software Maintenance ◽

Nonmonotonic Logic ◽

Research Progress ◽

Software Systems ◽

Practical Implementation ◽

Simple Extension ◽

Logic Programs ◽

Software Environment

This paper explores the possibility of automated support for detecting inconsistencies in software systems and requirements. The inconsistencies are introduced when the environment of the software system changes. We refer to the software environment as its context. We review the recent research progress on nonmonotonic logics, pointing out the significance of these results to software maintenance. We explain how a practical implementation of such logics can be obtained via a simple extension to logic programming in the form of an answer procedure that realizes the Extended Logic Semantics [7] for nonmonotonic logic programs that have a unique answer set (which is a large and useful class of logic programs). We augment the existing automated capabilities of the Computer-Aided Prototyping System (CAPS) for rapid prototyping via the extension to logic programming to provide an improved automated capability for detecting certain kinds of inconsistencies created by implicit requirements changes. We illustrate the significance of this capability via an example prototype for a problem originally suggested by Lehman.

Download Full-text