An Interval Type-2 Fuzzy Set Approach to Breast Cancer Dataset Analysis

The analysis of medical data is frequently characterized with uncertainties which tend to attract complexity. Therefore in this paper, an Interval Type-2 fuzzy set model: Hao and Mendel Approach (HMA) is proposed to fuzzify breast cancer data in order to handle quantitative attribute sharp boundary...

Full description

Saved in:
Bibliographic Details
Published inLecture notes in engineering and computer science Vol. 2241; p. 426
Main Authors Oladipupo, Olufunke, Olajide, Oluwole, Olawoye, Simisola, Oyelade, Jelili, Stephen, Adubi
Format Journal Article
LanguageEnglish
Published Hong Kong International Association of Engineers 22.10.2019
Subjects
Breast cancer
Computation
Datasets
Fuzzy sets
Intervals
Uncertainty
Online AccessGet full text

Cover

More Information
Summary:The analysis of medical data is frequently characterized with uncertainties which tend to attract complexity. Therefore in this paper, an Interval Type-2 fuzzy set model: Hao and Mendel Approach (HMA) is proposed to fuzzify breast cancer data in order to handle quantitative attribute sharp boundary problem and resolve inter and intra uncertainties. The HMA comprises of the data and the fuzzy part to create interval type-2 fuzzy values. The data part involves data preprocessing of the experts' intervals and the fuzzy set part establishes the structure of the FOU. The type reduction of the aggregated FOU is achieved by computing the centroid (measure of uncertainty) of the Fuzzy Set using the Enhanced Kernik-Mendel (EKM) approach. The defuzzification of the outcome which is an interval Type-2 Fuzzy set is achieved by computing the average of the interval's two endpoints; this captures and reflects the aggregate uncertainty of all the medical experts for breast cancer analysis. This will enhance interpretability of discrete intervals in medical dataset, providing a smooth transition from a fuzzy set to another in order to handle the sharp boundary interval problem and cater for inter and intra uncertainty in data interval value as the same word has diverse connotations to different people.
ISSN:2078-0958
2078-0966