¿Es posible aprender inductivamente de la experiencia?
- Arboleda-Quintero, Dairon Alberto; Patiño, Margarita E.
With the birth of probability calculus as a new risk quantification methodology, it also appears the debate with respect to the viability of inductive probability or to the fact that classic statistics uses inductive or deductive methods to draw its conclusions which is the basis of the scientific method. Such discussion becomes more intense with the strengthening of Bayesian statistics that seems to be the most promissory answer in favor of inductive learning and which is supported here by the authors of this article. There are many philosophers that reject the possibility to learn inductively from the experience. The most...
A system of logic, ratiocinative and inductive ; being a connected view of the principles of evidence and the methods of scientific investigation.
- Mill, John Stuart,
Mode of access: Internet.
Sistema de tareas docentes para el trabajo independiente en Medicina Natural y Tradicional
- Juviel Rodríguez, Matilde Norma; Trujillo Juviel, Pablo
Fundamento: el trabajo independiente es una modalidad de organización docente que contribuye a que los alumnos aprendan a estudiar con sus propios esfuerzos, estimulen su actividad creadora y formen hábitos correctos para la autosuperación.Objetivo: diseñar un sistema de tareas docentes para desarrollar habilidades de trabajo independiente a través de los contenidos de Medicina Natural y Tradicional.Métodos: se realizó una investigación de tipo descriptivo, en la Filial Universitaria de la Salud de Aguada de Pasajeros durante el curso 2012-2013. Se utilizaron como métodos teóricos: analítico-sintético, inductivo-deductivo e histórico-lógico; empíricos: análisis documental, observación, entrevistas a profesores y estudiantes, y para la valoración...
Necesidades de aprendizaje sobre rehabilitación bucomaxilofacial en residentes y especialistas de Neurocirugía
- Monteagudo Santiago, Janet; Ardisana Santana, Ernesto Fidel; Acosta Rodríguez, Juan Carlos
Fundamento: la prótesis bucomaxilofacial comprende la rehabilitación morfofuncional de las estructuras intrabucales y parabucales por medios artificiales.Objetivo: identificar las necesidades de aprendizaje y las insuficiencias en el desarrollo de habilidades sobre rehabilitación bucomaxilofacial en residentes y especialistas de Neurocirugía.Métodos: se realizó una investigación descriptiva transversal en el Hospital Universitario “Arnaldo Milián Castro” de Villa Clara, entre enero-mayo de 2014. Se utilizaron métodos teóricos: analítico-sintético, histórico-lógico e inductivo-deductivo; empíricos: análisis documental del programa de la especialidad de Neurocirugía y el plan de superación de los últimos 5 años, y encuesta en forma de cuestionario de respuestas breves al total de la...
FUZZY LOGIC AND GENETIC ALGORITHM FOR OPTIMISING THE APPROXIMATE MATCH OF RULES BASED ON BACKPROPAGATION NEURAL NETWORKS
- Jun Srisutapan; Boonserm Kijsirikul
This paper presents an application of Fuzzy Logic(FL) and Genetic Algorithm(GA) for improving the approximate match of first-order Inductive Logic Programming(ILP) rules that is based on Backpropagation Neural Networks(BNN). With the help of FL, the evaluation of the truth values of logic programs is more problem-sophisticated, before these values are sent to the BNN for learning or for recognising. We employ GA to find the best fuzzy sets. Experimental results on a Thai-OCR domain show that the our method gives the best recognition accuracy of 85.95 % compared to 82.31% recognition accuracy of the previous method. 1.
Programming An Inductive by Demonstration: Learning Formulation*
- Tessa A. Lau; Daniel S. Weld
Although Programming by Demonstration (PBD) has the potential to improve the productivity of unsophisticated users, previous PBD systems have used brittle, heuristic, domain-specific approaches to execution-trace generalization. In this paper we define two applicationindependent methods for performing generalization that are based on well-understood machine learning technology. TGENV ~ uses version-space generalization, and TGENFOIL is based on the FOIL inductive logic programming algorithm. We analyze each method both theoretically and empirically, arguing that TGENVS has lower sample complexity, but TGENFOIL can learn a much more interesting class of programs.
Incremental Learning of Functional Logic Programs
- C. Ferri-ramírez; J. Hernández-orallo; M. J. Ramírez-quintana
Abstract. Inthiswork,weconsidertheextensionoftheInductiveFunctional Logic Programming (IFLP) framework in order to learn functions in an incremental way. In general, incremental learning is necessary when the number of examples is infinite, very large or presented one by one. We have performed this extension in the FLIP system, an implementation of the IFLP framework. Several examples of programs which have been induced indicate that our extension pays off in practice. An experimental study of some parameters which affect this efficiency is performed and some applications for programming practice are illustrated, especially small classification problems and data-mining of semi-structured data.
Learning Functional Logic Classification Concepts From Databases
- C. Ferri-ramírez; J. Hernández-orallo; M. J. Ramírez-quintana
Abstract. In this paper we address the possibilities, advantages and shortcomings of addressing different data-mining problems with the InductiveFunctionalLogicProgramming(IFLP)paradigm.Asafunctional extension of the Inductive Logic Programming (ILP) approach, IFLP has all the advantages of the latter but the potential of a more natural representation language for classification, clustering and functional dependencies problems. Two issues are extremely important for successfully tackling these problems: incremental learning to handle large volumes of data and a consistent and flexible classes distribution evaluation to select among many possible hypotheses. We illustrate how these features are included in the IFLP paradigm and show some results with our...
A survey of (pseudo)-distance functions for structured data ∗ V. Estruch-Gregori C. Ferri-Ramírez J. Hernández-Orallo M.J. Ramírez-Quintana
- Departament Sistemes; Informàtics Computació
Learning from structured data is becoming increasingly important. Besides the well-known approaches which deals directly with complex data representation (inductive logic programming and multi relational data mining), recently new techniques have been proposed by upgrading propositional learning algorithms. Focusing on distance-based methods, they are extended by incorporating similarity functions defined over structured domains, for instance a k-NN algorithm solving a graph classification problem. Since a measure between objects is the essential component for this kind of methods, this paper consists of a brief survey about some of the recent similarity functions defined over common structured data (lists, sets, terms, etc.).
A Strong Complete Schema for Inductive Functional Logic Programming
- J. Hernández-orallo; M. J. Ramírez-quintana
Abstract. A new IFLP schema is presented as a general framework for theinductionoffunctionallogicprograms(FLP).Sincenarrowing(which is the most usual operational semantics of FLP) performs a unification (mgu) followed by a replacement, we introduce two main operators in our IFLP schema: a generalisation and an inverse replacement or intrareplacement, which results in a generic inversion of the transitive property of equality. We prove that this schema is strong complete in the way that, given some evidence, it is possible to induce any program which could have generated that evidence. We outline some possible restrictions in order to improve the tractability of the schema....
- Peter Geibel; Brijnesh J. Jain (eds; Cover Design; Kai-uwe Kühnberger; Peter König; Petra Ludewig; Peter Geibel; Brijnesh J. Jain; Thorsten Hinrichs
In recent years, machine learning approaches that operate on non-vectorial data like sequences, trees, graphs, and logical descriptions have gained increasing importance. This is primarily due to new exciting application areas like bioinformatics, text/web mining, and computer linguistics. In these fields, data is often represented by sequences, trees, and graphs of varying length. Examples are DNA molecules and proteins in bioinformatics, or words and documents in text mining. Recently, new powerful tools for solving classification tasks and other learning problems have been developed. Examples include string, tree, and graph kernels, learning in non-metric distance spaces, conditional random fields, learning in...
LEARNING SEMANTIC MODELS AND GRAMMAR RULES OF BUILDING PARTS
- Y. Dehbi; J. Schmittwilken; L. Plümer
Building reconstruction and building model generation nowadays receives more and more attention. In this context models such as formal grammars play a major role in 3D geometric modelling. Up to now, models have been designed manually by experts such as architects. Hence, this paper describes an Inductive Logic Programming (ILP) based approach for learning semantic models and grammar rules of buildings and their parts. Due to their complex structure and their important role as link between the building and its outside, straight stairs are presented as an example. ILP is introduced and applied as machine learning method. The learning process...
Negative robust learning results for horn clause programs
- Richard Nock; Olivier Gascuel
jappylIrmm.fr nocklirmm.fr gascuellirmm.fr low two different approaches to hypothesis production. MIS and CLINT, for instance, identify the target at We study the learn ability of Inductive Logic t~e limit, whereas most others use polynomial heuris-Programming (ILP) concept classes with re- tICS for concept induction. Consequently, these sysspect to robust-learning. We first investigate tems are generally efficient learners, but, to our knowlthe class of k-Horn clauses, and show that it edge, none can be formally shown to find the target is not learnable in that model. We prove this concept in polynomial time. using a reduction on which we impose as...
Applying MML to ILP ∗
- D. L. Dowe Cèsar Ferri; José Hernández-orallo; María José; Ramírez Quintana
In Inductive Logic Programming (ILP), since logic is a complete (universal) language, innitely many possible hypotheses are compatible (hence plausible) given the evidence. An intrinsic way of selecting the most convenient hypothesis from the set of possible theories is not only useful for model selection but it is also useful for guiding the search in the hypotheses space, as some ILP systems have done in the past. One selection/search criterion is to apply Occam's razor, i.e. to rst select/try the simplest hypotheses which cover the evidence. In order to do this, it is necessary to measure how simple a theory...
P.: Evaluation and validation of two approaches to user profiling
- F. Esposito; G. Semeraro; S. Ferilli; M. Degemmis; N. Di Mauro; T. M. A. Basile; P. Lops
Abstract. In the Internet era, huge amounts of data are available to everybody, in every place and at any moment. Searching for relevant information can be overwhelming, thus contributing to the user’s sense of information overload. Building systems for assisting users in this task is often complicated by the difficulty in articulating user interests in a structured form- a profile- to be used for searching. Machine learning methods offer a promising approach to solve this problem. Our research focuses on supervised methods for learning user profiles which are predictively accurate and comprehensible. The main goal of this paper is the...
kLog -- A Language for Logical and Relational Learning with Kernels
- Paolo Frasconi; Fabrizio Costa; Luc De Raedt; Kurt De Grave
kLog is a logical and relational language for kernel-based learning. It allows users to specify logical and relational learning problems at a high level in a declarative way. It builds on simple but powerful concepts: learning from interpretations, entity/relationship data modeling, logic programming and deductive databases (Prolog and Datalog), and graph kernels. kLog is a statistical relational learning system but unlike other statistical relational learning models, it does not represent a probability distribution directly. It is rather a kernel-based approach to learning that employs features derived from a grounded entity/relationship diagram. These features are derived using a novel technique called...
Function-free horn clauses are hard to approximate, in
- Richard Nock
Abstract. In this paper, we show two hardness results for approximating the best function-free Horn clause by an element of the same class. Our first result shows that for some constant k> 0, the error rate of the best k-Horn clause cannot be approximated in polynomial time to within any constant factor by an element of the same class. Our second result is much stronger. Under some frequently encountered complexity hypothesis, we show that if we replace the constant number of Horn clauses by a small, poly-logarithmic number, the constant factor blows up exponentially to a quasi-polynomial factor n l°gk...
Theorem Proving for Maude’s Rewriting Logic
- Èmes Al; Vlad Rusu; Manuel Clavel; Vlad Rusu; Manuel Clavel; Systèmes Communicants; Projet Vertecs
Publication interne n˚1873 — Novembre 2007 — 45 pages Abstract: We present an approach based on inductive theorem proving for verifying invariance properties of systems specified in Rewriting Logic, an executable specification language implemented (among others) in the Maude tool. Since theorem proving is not directly available for rewriting logic, we define an encoding of rewriting logic into its membership equational (sub)logic. Then, inductive theorem provers for membership equational logic, such as the itp tool, can be used for verifying the resulting membership equational logic specification, and, implicitly, for verifying invariance properties of the original rewriting logic specification. The approach...
Efficient Classification across Multiple Database Relations: A Crossmine Approach
- Xiaoxin Yin; Jiawei Han; Senior Member; Jiong Yang; Philip S. Yu
Abstract—Relational databases are the most popular repository for structured data, and is thus one of the richest sources of knowledge in the world. In a relational database, multiple relations are linked together via entity-relationship links. Multirelational classification is the procedure of building a classifier based on information stored in multiple relations and making predictions with it. Existing approaches of Inductive Logic Programming (recently, also known as Relational Mining) have proven effective with high accuracy in multirelational classification. Unfortunately, most of them suffer from scalability problems with regard to the number of relations in databases. In this paper, we propose a...
Using Multiple Clause Constructors in Inductive Logic Programming for Semantic Parsing
- Lappoon R. Tang; Raymond J. Mooney
In this paper, we explored a learning approach which combines different learning methods in inductive logic programming (ILP) to allow a learner to produce more expressive hypotheses than that of each individual learner. Such a learning approach may be useful when the performance of the task depends on solving a large amount of classification problems and each has its own characteristics which may or may not fit a particular learning method. The task of semantic parser acquisition in two different domains was attempted and preliminary results demonstrated that such an approach is promising.