Analysis of health habits of young people by extracting knowledge from database

Zbog globalne pandemije koronavirusa koja u trenutku pisanja rada vlada svijetom, prepoznata je važnost analize zdravstvenih navika mladog stanovništva Republike Hrvatske. S obzirom na to da lijek za navedeni virus nije otkriven, održavanje imuniteta i zdravlja organizma je dobilo sve veću važnost. Analiza je provedena korištenjem programa za rudarenje podataka Weka. Podaci za analizu su prikupljeni putem javno dostupne ankete, dok su većina pitanja preuzeta sa stranice Hrvatskog zavoda za javno zdravstvo. Osim analize zdravstvenih navika, provedena je klaster analiza, algoritmom k-srednjih vrijednosti, kako bi se pronašla povezanost između zdravstvenih navika i oboljenja od COVID-19 bolesti i na taj način pronašla moguća mjera zaštite od bolesti koja je napala svijet. Promatrani podaci su svrstani u deset grupa (klastera) gdje su se analizirali atributi klastera s klasom oboljelih od koronavirusa, kako bi se pronašao obrazac povezanosti.Due to the global coronavirus pandemic that rules the world at the time of writing, the importance of analyzing the health habits of the young population has been recognized. Since the cure for this virus has not been discovered, improving immunity and general health has become increasingly important. The analysis was performed using Weka’s data mining program. Data for analysis were collected through a publicly available questionnaire, while most questions were taken from the website of the Croatian Institute of Public Health. In addition to health habits analysis, a cluster analysis was performed using the kmeans algorithm to find the association between health habits and COVID-19 disease and thus find a possible measure of protection against the disease. The observed data were classified into ten groups (clusters) where the attributes of the cluster with the class of coronavirus patients were analyzed to find a pattern of association