41.
Learning Time Series Models with Inductive Logic Programming
- Alexessander Alves; Rui Camacho; Eugénio Oliveira
ABSTRACT: This paper reports on a set of proposals that make Inductive Logic Programming (ILP) systems adequate for inducing time series models. The proposals include an improvement in the ILP search process by the introduction of a statistical model validation step. We propose the definition of an adequate cost function based on the information criteria. The definition of the model evaluation step consists in an intuitive statistics that limits the minimum accepted performance of an induced hypothesis. The ILP system we used was provided with a library of background knowledge predicates adequate for time series problems. The proposals described in...
42.
Pharmacophore Discovery using the Inductive Logic Programming System Progol
- Paul Finn; David Page; Ronny Kohavi; Foster Provost
This paper is a case study of a machine aided knowledge discovery process within the general area of drug design. More specifically, the paper describes a sequence of experiments in which an Inductive Logic Programming(ILP) system is used for pharmacophore discovery. Within drug design, a pharmacophore is a description of the substructure of a ligand (a small molecule) which is responsible for medicinal activity. This medicinal activity is produced by interaction between the ligand and a binding site on a target protein. ILP was chosen by the domain expert (first author) at Pfizer since active molecules are most naturally described,...
43.
Inductive Logic Programming
- Stephen Muggleton,Glasgow G Ad
A new research area, Inductive Logic Programming, is presently
emerging. While inheriting various positive characteristics of the parent subjects
of Logic Programming and Machine Learning, it is hoped that the new area will
overcome many of the limitations of its forebears. The background to present
developments within this area is discussed and various goals and aspirations for
the increasing body of researchers are identified. Inductive Logic Programming
needs to be based on sound principles from both Logic and Statistics. On the
side of statistical justification of hypotheses we discuss the possible relationship between
Algorithmic Complexity theory and Probably-Approximately-Correct (PAC)
Learning. In terms of logic we provide a unifying framework...
44.
Discovering Rules to Design Newspapers: an Inductive Constraint Logic Programming Approach
- Marc Bernard; Francois Jacquenet; Avenue Alain Savary
Inductive Logic Programming (ILP) combines both Machine Learning and Logic Programming techniques. ILP uses first-order predicate logic restricted to Horn clauses as an underlying language. Thus, programs induced by an ILP system inherit the classical limitations of PROLOG programs. Constraint Logic Programming avoids some of the limitations of Logic Programming and so ILP aims to induce programs that employ this paradigm. Current ILP systems which induce constrained logic programs extend systems based on the normal semantics of ILP. In this paper we introduce ICLog, a new system which induces constrained logic programs and relies on an extension of a non-monotonic...
45.
Scaling up inductive logic programming by learning from interpretations. Data Mining and Knowledge Discovery
- Hendrik Blockeel; Nico Jacobs; Bart Demoen; Saˇso Dˇzeroski; Nada Lavrač
Abstract. When comparing inductive logic programming (ILP) and attribute-value learning techniques, there is a trade-off between expressive power and efficiency. Inductive logic programming techniques are typically more expressive but also less efficient. Therefore, the data sets handled by current inductive logic programming systems are small according to general standards within the data mining community. The main source of inefficiency lies in the assumption that several examples may be related to each other, so they cannot be handled independently. Within the learning from interpretations framework for inductive logic programming this assumption is unnecessary, which allows to scale up existing ILP algorithms....
46.
Inductive Logic Program Synthesis with
- Pierre Flener
DIALOGS (Dialog-based Inductive and Abductive LOGic program Synthesizer) is a schema-guided synthesizer
of recursive logic programs; it takes the initiative and minimally queries a (possibly computationally naive)
specifier for evidence in her/his conceptual language. The specifier must know the answers to such simple queries,
because otherwise s/he wouldn't even feel the need for the synthesized program. DIALOGS can be used by
any learner (including itself) that detects, or merely conjectures, the necessity of invention of a new predicate.
Due to its foundation on a powerful codification of a "recursion-theory" (by means of the template and constraints
of a divide-and-conquer schema), DIALOGS needs very little evidence and...
47.
Inductive Logic Program Synthesis with
- Pierre Flener
. dialogs (Dialogue-based Inductive and Abductive LOGic
program Synthesizer) is a schema-guided synthesizer of recursive logic
programs; it takes the initiative and queries a (possibly computationally
naive) specifier for evidence in her/his conceptual language. The specifier
must know the answers to such simple queries, because otherwise s/he
wouldn't even feel the need for the synthesized program. dialogs can be
used by any learner (including itself) that detects, or merely conjectures,
the necessity of invention of a new predicate. Due to its foundation on a
powerful codification of a "recursion-theory" (by means of the template
and constraints of a divide-and-conquer schema), dialogs needs very
little evidence and is very fast.
1 Introduction
This...
48.
Stochastic Inductive Logic Programming
- Ivan Bratko,Matevz Kovacic
Machine learning is an important part of artificial intelligence and its applications. Learning
from instances is one of the most active areas within machine learning. Initial successes
in the induction of propositional theories have been followed by algorithms that construct
hypotheses in the form of (a subset of) the first order relational concepts. Such learning is
called Inductive Logic Programming (ILP).
This thesis deals with two key problems of machine learning of concepts from instances:
hypothesis justification and hypothesis construction which are also a vital part of the form of
reasoning called inductive inference.
The purpose of concept formation is information compression (a hypothesis describes or
"explains" given learning...
49.
Inductive Logic Programming and Case-Based Reasoning for Nuclear Power Plants Monitoring
- Claire Nicolini; Cea Centre De Valduc; Francois Jacquenet; Marc Bernard
In this paper, we present the SECAPI system, a decision-aid tool in case of malfunction of nuclear power plants from the French nuclear power board (CEA). In case of malfunction of the plant, we do not have any formal specification of the running process. So it is not possible to implement some classical techniques such as Expert Systems. Machine learning techniques integrated in SECAPI, combine concepts of Case-Based Reasoning (CBR) and Inductive Logic Programming (ILP). The use of such a tool, allows to store past experiments which will be used later to explain new malfunctions and to induce rules that...
50.
Contributions to Inductive Logic Programming
- Afstudeerscriptie Bestuurlijke Informatica; R. M. De Wolf; What Is Inductive Logic Programming
Contents Preface iii 1 What is Inductive Logic Programming? 1 1.1 The importance of learning : : : : : : : : : : : : : : : : : : : : : 1 1.2 Inductive learning : : : : : : : : : : : : : : : : : : : : : : : : : : 2 1.3 The problem setting for ILP : : : : : : : : : : : : : : : : : : : : : 4 1.4 Other problem settings :...
51.
Advantages of Decision Lists and Implicit Negatives in Inductive Logic Programming
- Mary Elaine Califf; Raymond J. Mooney
This paper demonstrates the capabilities of Foidl, an inductive logic programming (ILP) system whose distinguishing characteristics are the ability to produce first-order decision lists, the use of an output completeness assumption to provide implicit negative examples, and the use of intensional background knowledge. The development of Foidl was originally motivated by the problem of learning to generate the past tense of English verbs; however, this paper demonstrates its superior performance on two different sets of benchmark ILP problems. Tests on the finite element mesh design problem show that Foidl's decision lists enable it to produce better results than all other...
52.
Inductive Constraint Logic
- Luc De Raedt,Wim Van Laer
. A novel approach to learning first order logic formulae from positive and negative examples is presented. Whereas present inductive logic programming systems employ examples as true and false ground facts (or clauses), we view examples as interpretations which are true or false for the target theory. This viewpoint allows to reconcile the inductive logic programming paradigm with classical attribute value learning in the sense that the latter is a special case of the former. Because of this property, we are able to adapt AQ and CN2 type algorithms in order to enable learning of full first order formulae. However,...
53.
Grammar Learning Using Inductive Logic
- Stephen Pulman,James Cussens
This paper gives a brief introduction to a particular machine learning method known as
inductive logic programming. It is argued that this method, unlike many current statistically
based machine learning methods, implies a view of grammar learning that bears close
a#nity to the views linguists have of the `logical problem of language acquisition'.
54.
Query Optimization in Inductive Logic
- Jan Struyf,Hendrik Blockeel
Query optimization is used frequently in relational database
management systems. Most existing techniques are based on reordering
the relational operators, where the most selective operators are executed
first. In this work we evaluate a similar approach in the context of Inductive
Logic Programming (ILP). There are some important di#erences
between relational database management systems and ILP systems. We
describe some of these di#erences and list the resulting requirements for
a reordering transformation suitable for ILP. We propose a transformation
that meets these requirements and an algorithm for estimating the
computational cost of literals, which is required by the transformation.
55.
Bases in Inductive Logic Programming
- Maurice Bruynooghe,Sofie Verbaeten,Hendrik Blockeel
In many applications of Inductive Logic Programming (ILP), learning
occurs from a knowledge base that contains a large number of examples.
56.
Parallel Inductive Logic for Data Mining
- Yu Wang,David Skillicorn
Data mining is the process of automatic extraction of novel, useful and understandable patterns in
very large databases. High-performance, scalable, and parallel computing algorithms are crucial in data
mining as datasets grow in size and complexity. Inductive logic is a research area in the intersection of
machine learning and logic programming, which has been recently applied to data mining. Inductive
logic studies learning from examples, within the framework provided by clausal logic. It provides a
uniform and expressive means of representation: examples, background knowledge, and induced theories
are all expressed in first-order logic. Such an expressive representation is computationally expensive,
so it is natural to consider improving...
57.
Web Usage Mining with Inductive Logic Programming
- Amund Tveit
This paper suggests an experimental approach of how to apply inductive logic programming in the discovery of web usage patterns in the form of first-order rules representing user sessions. Such rules may be used to improve the quality and the performance of a web site. The experiment has been done using the Progol Inductive Logic Programming System, and the data source are log files from an Apache web server.
58.
A Logic of Non-Monotone Inductive De
- Marc Denecker
Well-known principles of induction include monotone induction
and dierent sorts of non-monotone induction such as inationary
induction, induction over well-ordered sets and iterated induction. In
this work, we de
ne a logic formalizing induction over well-ordered sets
and monotone and iterated induction. Just as the principle of positive
induction has been formalized in FO(LFP), and the principle of inationary
induction has been formalized in FO(IFP), this paper formalizes
the principle of iterated induction in a new logic for Non-Monotone Inductive
De
nitions (NMID-logic). The semantics of the logic is strongly
inuenced by the well-founded semantics of logic programming.
59.
Parallel Inductive Logic in Data Mining
- Yu Wang
Data-mining is the process of automatic extraction of novel, useful and understandable patterns from very large
databases. High-performance, scalable, and parallel computing algorithms are crucial in data mining as datasets
grow inexorably in size and complexity. Inductive logic is a research area in the intersection of machine learning
and logic programming, which has been recently applied to data mining. Inductive logic studies learning from
examples, within the framework provided by clausal logic. It provides a uniform and very expressive means
of representation: All examples, background knowledge as well as the induced theory are expressed in first-order
logic. However, such an expressive representation is often computationally expensive....
60.
Representing Biases for Inductive Logic Programming
- Birgit Tausend
. As each of the four main approaches to a declarative bias
represention in Inductive Logic Programming (ILP), the representation
by parameterized languages or by clause sets, the grammar-based and
the scheme-based representation, fails in representing all language biases
in ILP systems, we present a unifying representation language MILESCTL
for these biases by extending the scheme-based approach.
1 Introduction
Describing an inductive learning system at the knowledge level [Die86] requires
to characterize the amount of new knowledge added by induction. As the hypotheses
vary with respect to the bias, characterizing the new knowledge means
to describe the bias. Among others, the bias involves the hypothesis language.
For example, the language bias...
