Information-Based Selection of Abstraction Levels
نویسنده
چکیده
We model the learning of classifications as a combination of abstraction and class assignment. We discuss the problem of selecting the most suitable of multiple abstractions for this purpose. Weaker abstractions perform better on training sets, but typically do not generalize very well. Stronger abstractions often generalize better, but may fail to include important properties. We introduce the relative information gain as a criterion to determine an optimal balance between precision and generality of abstractions. Experimental results with abstractions used for the classification of terms indicate the success of this approach.
منابع مشابه
Video Abstraction in H.264/AVC Compressed Domain
Video abstraction allows searching, browsing and evaluating videos only by accessing the useful contents. Most of the studies are using pixel domain, which requires the decoding process and needs more time and process consuming than compressed domain video abstraction. In this paper, we present a new video abstraction method in H.264/AVC compressed domain, AVAIF. The method is based on the norm...
متن کاملA New Model for Best Customer Segment Selection Using Fuzzy TOPSIS Based on Shannon Entropy
In today’s competitive market, for a business firm to win higher profit among its rivals, it is of necessity to evaluate, and rank its potential customer segments to improve its Customer Relationship Management (CRM). This brings the importance of having more efficient decision making methods considering the current fast growing information era. These decisions usually involve several criteria,...
متن کاملA New Hybrid Framework for Filter based Feature Selection using Information Gain and Symmetric Uncertainty (TECHNICAL NOTE)
Feature selection is a pre-processing technique used for eliminating the irrelevant and redundant features which results in enhancing the performance of the classifiers. When a dataset contains more irrelevant and redundant features, it fails to increase the accuracy and also reduces the performance of the classifiers. To avoid them, this paper presents a new hybrid feature selection method usi...
متن کاملToward Automatic Test Pattern Generation for VHDL Descriptions
This paper describes an approach for defining a model for the VHDL descriptions which can be used for test generation purpose. The VHDL description can be transformed to this model by semantic preserving transformations without lost of information needed for test generation purpose. Together with the model definition a unified fault model is defined which can be easily related to well known fau...
متن کاملEnvironmental Planning for Wind Power Plant Site Selection using a Fuzzy PROMETHEE-Based Outranking Method in Geographical Information System
Selection of suitable sites for wind power plants is one of the most important decision on wind resources development. Site selection for the establishment of large wind power plants requires spatial evaluation taking technical, economic, and environmental considerations into account. This study has applied a combination of PROMETHEE and Fuzzy AHP methods in a geographical information system en...
متن کامل