| Popis: |
U radu je provedena analiza podataka na povijesnim podacima o utrkama Formule 1. Na temelju analize su izvedeni zaključci o značajkama koje bi mogle utjecati na vjerojatnost pobjede vozača. Odabrane su najbitnije značajke i kombinirane u skup podataka koji je korišten za treniranje modela. Problem je tretiran i kao klasifikacijski i kao regresijski pa su korištene prikladne metode strojnog učenja. Prikazani su i uspoređeni rezultati korištenih metoda. In the paper, data analysis was performed on historical data on Formula 1 races. Based on the research, conclusions were drawn about the features that could affect the probability of a driver's victory. The most important features were selected and combined into a dataset that was used to train the model. The problem was treated as both classification and regression, so appropriate machine learning methods were used. The results of the methods used are presented and compared. |
