Learning family relationships exploiting multistrategy
نویسنده
چکیده
This work presents the application of INTHELEX, an incremental learning system enhanced by multistrategy capabilities, on a dataset concerning family relationships. The aim is to investigate if and how much abduction, abstraction and deduction can support pure induction. The reported experimental results, although preliminary, demonstrate that such a cooperation may improve efficiency and effectiveness of the learning process.
منابع مشابه
Multistrategy Learning: When, How and Why
In this paper we describe our experience in designing, implementing and using multistrategy approaches to machine learning. In particular, the system SMART+, WHY and REGAL, each one interpreting "multistrategy" in a different way, will be briefly described, and the lessons learned from their realization commented upon.
متن کاملAn Empirical Study of Computational Introspection: Evaluating Introspective Multistrategy Learning in the Meta-AQUA Syste
The theory of introspective multistrategy learning proposes that three transformations must occur to learn effectively from a performance failure in an intelfigent system: Blame assignment, deciding what to learn, and learning-strategy construction. The Meta-AQUA system is a multistrategy learner that operates in the domain of story-understanding failures and is designed to evaluate this learni...
متن کاملMacro and micro perspectives of multistrategy learning
Machine learning techniques are perceived to have a great potential as means for the acquisition of knowledge; nevertheless, their use in complex engineering domains is still rare. Most machine learning techniques have been studied in the context of knowledge acquisition for well defined tasks, such as classification. Learning for these tasks can be handled by relatively simple algorithms. Comp...
متن کاملIn Machine Learning: a Multistrategy Approach, Vol. Iv Macro and Micro Perspectives of Multistrategy Learning
Machine learning techniques are perceived to have a great potential as means for the acquisition of knowledge; nevertheless, their use in complex engineering domains is still rare. Most machine learning techniques have been studied in the context of knowledge acquisition for well deened tasks, such as classiication. Learning for these tasks can be handled by relatively simple algorithms. Comple...
متن کاملText Categorization through Multistrategy Learning and Visualization
This paper introduces a multistrategy learning approach to the categorization of text documents. The approach benefits from two existing, and in our view complimentary, sets of categorization techniques: those based on Rocchios algorithm and those belonging to the rule learning class of machine learning algorithms. Visualization is used for the presentation of the output of learning.
متن کامل