Ανάλυση κλινικών δεδομένων με χρήση αλγορίθμων εξόρυξης δεδομένων για την πρόβλεψη του διαβήτη
View/ Open
Keywords
ΔιαβήτηςAbstract
According to statistics, studies and the development of medicine with new machines and
technological tools that daily contribute to a better quality of life and immediate healing of
patients, it is proven that machine learning has brought significant developments in the field
of health and medicine. It has been applied in many areas to improve diagnosis, prevention,
treatment and healthcare delivery. Among the applications of machine learning in healthcare
include personalised therapy i.e. the analysis of genetic data to create personalised
treatments and medicines for specific patient cases, health management which can lead to
optimal management of hospital resources as well as digital health and healthcare providers
such as medical records management. This places it in a very, very important and key area
that we increasingly want as a society to develop and evolve.
This thesis is about the comparative study of machine learning algorithms and the evaluation
of their performance for diabetes diagnosis using the Diabetes prediction dataset clinical
dataset.
For each of the machine learning models, we analyzed several metrics to conclude which one
provided the best result for our dataset.