Rank thresholds in classifier ensembles in medical diagnosis
Journal Title: Computer Science and Mathematical Modelling - Year 2016, Vol 0, Issue 3
Abstract
Classification methods have multiple applications, with medical diagnosis being one of the most common. A powerful way to improve classification quality is to combine single classifiers into an ensemble. One of the approaches for creating such ensembles is to combine class rankings from base classifiers. In this paper, two rank-based ensemble methods are studied: Highest Rank and Borda Count. Furthermore, the effect of applying class rank threshold to these methods is analyzed. We performed tests using real-life medical data. It turns out that specificity of data domain can affect classification quality depending on classifier type.
Authors and Affiliations
Karol Antczak
Keywords
Related Articles
- EP ID EP186316
- DOI 10.5604/01.3001.0009.4412
- Views 100
- Downloads 0
Uproszczona metoda wyznaczania gier równoważnych i ich rozwiązań
W pracy zaproponowano nowe charakterystyki wieloosobowej gry kooperacyjnej, dotyczące możliwości negocjacyjnych graczy oraz ich koalicji. Zdefiniowano funkcję korzyści kooperacyjnych gry , która pozwala na szczegółową a...
Data Visualization While Determining Similarities of Medical Patterns
The article presents the concept of using the theory of similarity in the recognition of medical patterns. The aim of the work is to construct a graphical model of disease entity pattern and the state of the patient's he...
Assessment of ranking algorithms in complex networks
A particularly helpful search of a network such as the Internet or a citation network not only finds nodes that satisfy some criteria but also ranks those nodes for importance to create what amounts to a “reading list”....
Hidden associations in semantic networks
This paper presents main concepts of semantic network applications in facts association allowing for potential crisis situation identification. Described method helps with development of ontology centric tools for terror...
The system dynamics approach to combat modeling