Skip to Main Content
 

Global Search Box

 
 
 
 

Files

ETD Abstract Container

Abstract Header

Binary Classification With First Phase Feature Selection forGene Expression Survival Data

Abstract Details

2019, Master of Science, Ohio State University, Biostatistics.
We look at a variety of binary class models for feature selection and classification with deterministically censored survival data. We compare the discrimination and ability to select appropriate features to the predominant methods used for time to event analysis. Survival methods make a number of assumptions while building the hazard function that might be violated when dealing with real data. In situations where this happens, it might be worthwhile to look at methods that do not succumb to these same problems. While it is possible to construct more complex hazard functions, in some applications we may only want to look at chances of survival up to a predetermined future time point, and only output a statistic that depicts the probability of survival up to this predetermined follow up time. In such situations binary classification seems to be a promising methodology. Here we investigate binary classification for these special circumstances. Via simulations and analyzing 3 real datasets on cancer survivorship.
Grzegorz Rempala (Advisor)
Chi Song (Committee Member)
88 p.

Recommended Citations

Citations

  • Loveless, I. (2019). Binary Classification With First Phase Feature Selection forGene Expression Survival Data [Master's thesis, Ohio State University]. OhioLINK Electronic Theses and Dissertations Center. http://rave.ohiolink.edu/etdc/view?acc_num=osu1555444873531262

    APA Style (7th edition)

  • Loveless, Ian. Binary Classification With First Phase Feature Selection forGene Expression Survival Data. 2019. Ohio State University, Master's thesis. OhioLINK Electronic Theses and Dissertations Center, http://rave.ohiolink.edu/etdc/view?acc_num=osu1555444873531262.

    MLA Style (8th edition)

  • Loveless, Ian. "Binary Classification With First Phase Feature Selection forGene Expression Survival Data." Master's thesis, Ohio State University, 2019. http://rave.ohiolink.edu/etdc/view?acc_num=osu1555444873531262

    Chicago Manual of Style (17th edition)