| Popis: |
Several models have been developed for the forecasting of bankruptcy dataset but still, this is an active research area without which it may lead to a severe financial crisis. This paper focuses on the preprocessing phase which is very much essential in this domain to enhance the performance of the prediction model. We have filled out the missing values of the feature by the means of corresponding feature vector and then used oversampling technique SMOTE, normalization in the preprocessing phase is then applied to the transformed data set to five different popular classifiers such as random forest, decision tree, K-nearest neighbor (K-NN), logistic regression, and artificial neural network (ANN) to see the effect of these preprocessing steps in the prediction performance of all these five classifiers. |
