Binary classification techniques: An application on simulated and real bio-medical data

Fragkiskos G. Bersimis, Iraklis Varlamis, Malvina Vamvakari, Demosthenes B. Panagiotakos

Research output: A Conference proceeding or a Chapter in BookChapterpeer-review


This chapter investigates the performance of classification techniques for discrete variables associated with binomial outcomes. It presents various classification techniques based on multivariate indices and on machine learning methods, and evaluates their distinctive ability by using simulated data as well as real Greek medical data. The chapter assesses classification techniques by using criteria such as the area under the receiver operating characteristic curve, sensitivity and specificity. It evaluates the classification techniques’ predictability as well as their results’ statistical significance by using Monte Carlo cross-validation. Conclusively, the chapter proposes methods for the selection of an effective diagnostic method by using suitable classification methods or weighted indices in relation to the health data nature such as those derived from psychological diseases, nutritional adequacy and so on.

Original languageEnglish
Title of host publicationData Analysis and Applications 3
Subtitle of host publicationComputational, Classification, Financial, Statistical and Stochastic Methods
Place of PublicationUnited Kingdom
Number of pages31
ISBN (Electronic)9781119721871
ISBN (Print)9781786305343
Publication statusPublished - 18 Apr 2020
Externally publishedYes


Dive into the research topics of 'Binary classification techniques: An application on simulated and real bio-medical data'. Together they form a unique fingerprint.

Cite this