An attempt is made to determine how machine learning benefit from using a richer and more expressive representation. The approach taken is to compare the performance of a so-called upgrade of a propositional algorithm...
详细信息
An attempt is made to determine how machine learning benefit from using a richer and more expressive representation. The approach taken is to compare the performance of a so-called upgrade of a propositional algorithm and the performance of a propositional learning algorithm applied to so-called propositionalization. This comparison is performed in learning domains that are essentially relational.
inductive logic programming (ILP) is concerned with the induction of logic programs from examples and background knowledge. In ILP, the shift of attention from program synthesis to knowledge discovery resulted in adva...
详细信息
inductive logic programming (ILP) is concerned with the induction of logic programs from examples and background knowledge. In ILP, the shift of attention from program synthesis to knowledge discovery resulted in advanced techniques that are practically applicable for discovering knowledge in relational databases. This paper gives a brief introduction to ILP, presents selected ILP techniques for relational knowledge discovery and reviews selected ILP applications.
inductive logic programming (ILP) is the area of AI which deals with the induction of hypothesised predicate definitions from examples and background knowledge. logic programs are used as a single representation for e...
详细信息
inductive logic programming (ILP) is the area of AI which deals with the induction of hypothesised predicate definitions from examples and background knowledge. logic programs are used as a single representation for examples, background knowledge and hypotheses. ILP is differentiated from most other forms of Machine Learning (ML) both by its use of an expressive representation language and its ability to make use of logically encoded background knowledge. This has allowed successful applications of ILP in areas such as molecular biology and natural language which both have rich sources of background knowledge and both benefit from the use of an expressive concept representation languages. For instance, the ILP system Progol has recently been used to generate comprehensible descriptions of the 23 most populated fold classes of proteins, where no such descriptions had previously been formulated manually. In the natural language area ILP has not only been shown to have higher accuracies than various other ML approaches in learning the past tense of English but also shown to be capable of learning accurate grammars which translate sentences into deductive database queries. The area of Learning Language in logic (LLL) is producing a number of challenges to existing ILP theory and implementations. In particular, language applications of ILP require revision and extension of a hierarchically defined set of predicates in which the examples are typically only provided for predicates at the top of the hierarchy. New predicates often need to be invented, and complex recursion is usually involved. Advances in ILP theory and implementation related to the challenges of LLL are already producing beneficial advances in other sequence-oriented applications of ILP. In addition LLL is starting to develop its own character as a sub-discipline of AI involving the confluence of computational linguistics, machine learning and logicprogramming. (C) 1999 Elsevier Science B.V. All rights re
Hypotheses constructed by inductive logic programming (ILP) systems are finite sets of definite clauses. Top-down ILP systems usually adopt the following greedy clause-at-a-time strategy to construct such a hypothesis...
详细信息
Hypotheses constructed by inductive logic programming (ILP) systems are finite sets of definite clauses. Top-down ILP systems usually adopt the following greedy clause-at-a-time strategy to construct such a hypothesis: start with the empty set of clauses and repeatedly add the clause that most improves the quality of the set. This paper formulates and analyses an alternative method for constructing hypotheses. The method, called cautious induction, consists of a first stage, which finds a finite set of candidate clauses, and a second stage, which selects a finite subset of these clauses to form a hypothesis. By using a less greedy method in the second stage, cautious induction can find hypotheses of higher quality than can be found with a clause-at-a-time algorithm. We have implemented a top-down, cautious ILP system called CILS. This paper presents CILS and compares it to Progol, a top-down clause-at-a-time ILP system. The sizes of the search spaces confronted by the two systems are analysed and an experiment examines their performance on a series of mutagenesis learning problems.
Knowledge acquisition with machine learning techniques is a fundamental re-quirement for knowledge discovery from databases and data mining systems. Two techniquesin particular - inductive learning and theory revision...
详细信息
Knowledge acquisition with machine learning techniques is a fundamental re-quirement for knowledge discovery from databases and data mining systems. Two techniquesin particular - inductive learning and theory revision - have been used toward this end. Amethod that combines both approaches to effectively acquire theories (regularity) from a setof training examples is presented. inductive learning is used to acquire new regularity fromthe training examples; and theory revision is used to improve an initial theory. In addition, atheory preference criterion that is a combination of the MDL-based heuristic and the Laplaceestimate has been successfully employed in the selection of the promising theory. The resultingalgorithm developed by integrating inductive learning and theory revision and using the criterionhas the ability to deal with complex problems, obtaining useful theories in terms of its predictiveaccuracy.
The paper studies the learnability of Horn expressions within the framework of learning from entailment, where the goal is to exactly identify some pre-fixed and unknown expression by making queries to membership and ...
详细信息
The paper studies the learnability of Horn expressions within the framework of learning from entailment, where the goal is to exactly identify some pre-fixed and unknown expression by making queries to membership and equivalence oracles. It is shown that a class that includes both range restricted Horn expressions (where terms in the conclusion also appear in the condition of a Horn clause) and constrained Horn expressions (where terms in the condition also appear in the conclusion of a Horn clause) is learnable. This extends previous results by showing that a larger class is learnable with better complexity bounds. A further improvement in the number of queries is obtained when considering the class of Horn expressions with inequalities on all syntactically distinct terms. (C) 2002 Elsevier Science (USA).
作者:
Yamamoto, AHokkaido Univ
Div Elect & Informat Engn Kita Ku Sapporo Hokkaido 0608628 Japan Hokkaido Univ
Meme Media Lab Kita Ku Sapporo Hokkaido 0608628 Japan
In this paper we revise Muggleton's theory of inverse entailment, which is the logical foundation of Progol, one of the most famous ILP systems. We first point out that the theory is incomplete in general. Secondl...
详细信息
In this paper we revise Muggleton's theory of inverse entailment, which is the logical foundation of Progol, one of the most famous ILP systems. We first point out that the theory is incomplete in general. Secondly we prove that the theory is complete if the background knowledge given to the system is a ground reduced program, every training example is a ground unit clause, and the hypothesis space is the set of all definite clauses. The proof is obtained by showing that every ground reduced logic program is logically equivalent to the conjunction of all atoms in its least Herbrand model. As a corollary to this equivalence, we are finally able to improve the logical foundation of the GOLEM system.
Instance based learning and clustering are popular methods in propositional machine learning. Both methods use a notion of similarity between objects. This dissertation investigates these methods in a relational setti...
详细信息
Instance based learning and clustering are popular methods in propositional machine learning. Both methods use a notion of similarity between objects. This dissertation investigates these methods in a relational setting. First, a number of new metrics are proposed. Next, these metrics are used to upgrade clustering and instance based learning to first order logic.
The bounded ILP-consistency problem for function-free Horn clauses is described as follows. Given a set E+ and E- of function-free ground Horn clauses and an integer I;polynomial in E+ boolean OR E-, does there exist ...
详细信息
The bounded ILP-consistency problem for function-free Horn clauses is described as follows. Given a set E+ and E- of function-free ground Horn clauses and an integer I;polynomial in E+ boolean OR E-, does there exist a function-free Horn clause C with no more than k literals such that C subsumes each element in E+ and C does not subsume any element in E-? It is shown that this problem is Sigma(2)(P) complete. We derive some related results on the complexity of ILP and discuss the usefulness of such complexity results.
暂无评论