this paper presents a method for detecting the connecting points in connected thai printed characters. In thai Optical Character Recognition systems, an important problem that decreases the accuracy is the connected c...
详细信息
ISBN:
(纸本)0769519482
this paper presents a method for detecting the connecting points in connected thai printed characters. In thai Optical Character Recognition systems, an important problem that decreases the accuracy is the connected characters. these characters could cause the errors in segmentation process. To attack this problem, we first extract the features of the connecting points in the character images. then, we employ inductivelogicprogramming to produce the rules that will be used to classify the unseen images. Finally, we use a Backpropagation Neural Network to make these rules more flexible. the experimental results show that our method achieves 94.94% accuracy.
the goals of this presentation are as follows: - Review some key ideas and developments in inductivelogicprogramming. - Show how these ideas can be used in other learning settings, and in particular for the computat...
详细信息
ISBN:
(纸本)3540005676
the goals of this presentation are as follows: - Review some key ideas and developments in inductivelogicprogramming. - Show how these ideas can be used in other learning settings, and in particular for the computational scientific discovery of quantitative laws. - Encourage more research on learning in rich representations, such as relational representations and differential equations, which can be used for modeling a variety of real world problems.
Incremental learning from noisy data is a difficult task and has received very little attention in the field of inductivelogicprogramming. this paper outlines an approach to noisy incremental learning based on a pos...
详细信息
ISBN:
(纸本)3540005676
Incremental learning from noisy data is a difficult task and has received very little attention in the field of inductivelogicprogramming. this paper outlines an approach to noisy incremental learning based on a possible worlds model and its implementation in NILE. Several issues relating to the use of this model are addressed., Empirical results are shown for an existing batch domain and also for an interactive learning task.
inductive learning has been employed successfully in various domains, however the inductivelogicprogramming (ILP) systems focused on non-incremental learning tasks where independent sets of data are provided incoher...
详细信息
ISBN:
(纸本)9781538616390
inductive learning has been employed successfully in various domains, however the inductivelogicprogramming (ILP) systems focused on non-incremental learning tasks where independent sets of data are provided incoherently. In this paper, we propose a new genetic algorithm-based ILP system, called GAILP, for incremental learning. GAILP is a covering algorithm which extracts hypotheses/rules from a collection of examples in a reliable way. It employs a genetic algorithm technique to discover various aspects of the potential combinations. GAILP induces every possible rule for the given combination and selects the most generic ones among them. It also eliminates rules which might become obsolete by the existence of more generic rules. Unlike other ILP systems, GAILP batches all given examples and background knowledge, then it groups the examples and prioritizes the induction process. this prioritization needs to be done to preserve dependency and to revise theory. the paper introduces GAILP's fundamentals mechanisms and demonstrates its algorithms with a running example.
Two common characteristics of relational data sets-concentrated linkage and relational auto-correlation-can cause traditional methods of evaluation to greatly overestimate the accuracy of induced models on test sets. ...
详细信息
ISBN:
(纸本)3540005676
Two common characteristics of relational data sets-concentrated linkage and relational auto-correlation-can cause traditional methods of evaluation to greatly overestimate the accuracy of induced models on test sets. We identify these characteristics, define quantitative measures of their severity, and explain how they produce this bias. We show how linkage and autocorrelation affect estimates of model accuracy by applying FOIL to synthetic data and to data drawn from the Internet Movie Database. We show how a modified sampling procedure can eliminate the bias.
New representation languages that integrate first order logic with Bayesian networks have been proposed in the literature. Probabilistic Relational models (PRM) and Bayesian logic Programs (BLP) are examples. Algorith...
详细信息
ISBN:
(纸本)3540005676
New representation languages that integrate first order logic with Bayesian networks have been proposed in the literature. Probabilistic Relational models (PRM) and Bayesian logic Programs (BLP) are examples. Algorithms to learn boththe qualitative and the quantitative components of these languages have been developed. Recently, we have developed an algorithm to revise a BLP. In this paper, we discuss the relationship among these approaches, extend our revision algorithm to return the highest probabilistic scoring BLP and argue that for a classification task our approach, which uses techniques of theory revision and so searches a smaller hypotheses space, can be a more adequate choice.
An Event Calculus program to control the navigation of a real robot was generated using theory Completion techniques. this is an application of ILP in the non-observational predicate learning setting. this work utiliz...
详细信息
ISBN:
(纸本)3540005676
An Event Calculus program to control the navigation of a real robot was generated using theory Completion techniques. this is an application of ILP in the non-observational predicate learning setting. this work utilized 1) extraction-case abduction;2) the simultaneous completion of two, mutually related predicates;and 3) positive observations only learning. Given time-trace observations of a robot successfully navigating a model office and other background information, theory Completion was used to induce navigation control programs in the event calculus. Such programs consisted of many clauses (up to 15) in two mutually related predicates. this application demonstrates that abduction and induction can be combined to effect non-observational multi-predicate learning.
Previous papers have studied learning of Stochastic logic Programs (SLPs) either as a purely parametric estimation problem or separated structure learning and parameter estimation into separate phases. In this paper w...
详细信息
ISBN:
(纸本)3540005676
Previous papers have studied learning of Stochastic logic Programs (SLPs) either as a purely parametric estimation problem or separated structure learning and parameter estimation into separate phases. In this paper we consider ways in which boththe structure and the parameters of an SLP can be learned simultaneously. the paper assumes an ILP algorithm, such as Progol or FOIL, in which clauses are constructed independently. We derive analytical and numerical methods for efficient computation of the optimal probability parameters for a single clause choice within such a search.
In this paper we describe a new approach to the application of evolutionary stochastic search in inductivelogicprogramming (ILP). Unlike traditional approaches that focus on evolving populations of logical clauses, ...
详细信息
ISBN:
(纸本)9781450300728
In this paper we describe a new approach to the application of evolutionary stochastic search in inductivelogicprogramming (ILP). Unlike traditional approaches that focus on evolving populations of logical clauses, our refinement-based approach uses the stochastic optimization process to iteratively adapt initial working clause. Utilization of context-sensitive concept refinements (adaptations) helps the search operations to produce mostly syntactically correct concepts and enables using available background knowledge both for efficiently restricting the search space and for directing the search. thereby, the search is more flexible, less problem-specific and the framework can be easily used with any stochastic search algorithm within ILP domain. Experimental results on several data sets verify the usefulness of this approach.
the application of Genetic programming (GP) to the discovery of empirical laws most often suffers from two limitations. the first one is the size of the search space;the second one is the growth of non-coding segments...
详细信息
ISBN:
(纸本)3540005676
the application of Genetic programming (GP) to the discovery of empirical laws most often suffers from two limitations. the first one is the size of the search space;the second one is the growth of non-coding segments, the introns, which exhausts the memory resources as GP evolution proceeds. these limitations are addressed by combining Genetic programming and Stochastic Grammars. On one hand, grammars are used to represent prior knowledge;for instance, context-free grammars can be used to enforce the discovery of dimensionally consistent laws, thereby significantly restricting GP search space. On the other hand, in the spirit of distribution estimation algorithms, the grammar is enriched with derivation probabilities. By exploiting such probabilities, GP avoids the intron phenomenon. the approach is illustrated on a real-world like problem, the identification of behavioral laws in Mechanics.
暂无评论