On the Induction of Intelligible Ensembles
نویسنده
چکیده
Ensembles of classiiers, e.g. decision trees, often exhibit greater predictive accuracy than single classiiers alone. Bagging and boosting are two standard ways of generating and combining multiple classiiers. Unfortunately , the increase in predictive performance is usually linked to a dramatic decrease in intelligibility: ensembles are more or less black boxes comparable to neural networks. So far attempts at pruning of ensembles have not been very successful, approximately reducing ensembles into half. This paper describes a diierent approach which both tries to keep ensemble-sizes small during induction already and also limits the complexity of single classiiers rigorously. Single classiiers are decision-stumps of a prespeciied maximal depth. They are combined by majority voting. Ensembles are induced and pruned by a simple hill-climbing procedure. These ensembles can reasonably be transformed into equivalent decision trees. We conduct some empirical evaluation to investigate both predictive accuracies and classiier complexities.
منابع مشابه
Coherent Transport of Single Photon in a Quantum Super-cavity with Mirrors Composed of Λ-Type Three-level Atomic Ensembles
In this paper, we study the coherent transport of single photon in a coupled resonator waveguide (CRW) where two threelevel Λ-type atomic ensembles are embedded in two separate cavities. We show that it is possible to control the photon transmission and reflection coefficients by using classical control fields. In particular, we find that the total photon transmission and reflection are achieva...
متن کاملA High-Performance Model based on Ensembles for Twitter Sentiment Classification
Background and Objectives: Twitter Sentiment Classification is one of the most popular fields in information retrieval and text mining. Millions of people of the world intensity use social networks like Twitter. It supports users to publish tweets to tell what they are thinking about topics. There are numerous web sites built on the Internet presenting Twitter. The user can enter a sentiment ta...
متن کاملSurvey and Critique of First Intelligible and Secondary Philosoph ical In telligible in Transcendental Philosophy
متن کامل
Evaluating Unsupervised Ensembles when applied to Word Sense Induction
Ensembles combine knowledge from distinct machine learning approaches into a general flexible system. While supervised ensembles frequently show great benefit, unsupervised ensembles prove to be more challenging. We propose evaluating various unsupervised ensembles when applied to the unsupervised task of Word Sense Induction with a framework for combining diverse feature spaces and clustering ...
متن کاملConditional Equilibrium and the Equivalence of Microcanonical and Grandcanonical Ensembles in the Thermodynamic Limit
Equivalence (allowing for convex combinations) of microcanonical, canonical and grandcanonical ensembles for states of classical systems is established under very mild assumptions on the limiting state. We introduce the notion of conditional equilibrium (C.E.), a property of states of infinite systems which characterizes convex combinations of limits of microcanonical ensembles. It is shown tha...
متن کامل