An Alternative Approach to Causes of Death Prediction Using Support Vector Machines

We use Support Vector Machines, machine learning algorithms that have become very popular among classification problems, to classify mortality rates into specific macro classes of death causes. After dividing data into training and test data, the algorithm learns from the labeled training set to predict the cause-of-death label. Then, it automatically assigns the cause-of-death labels to the test dataset. The proposed model is formulated by exploiting the features of mortality dynamics such as age, gender, and central death rate and could be used to classify data for countries where causes of death are misreported. Furthermore, it is worth noting that the classification of causes of death has considerably changed since 1959, passing from ICD 7th to ICD 10th revision. Our model might be used to reconcile different causes-of-death classifications or classify residual classes in cause-of-death databases (e.g., in the WHO database, an archive of causes-of-death information for several countries worldwide).