What are the steps involved if it is possible to add two confusion matrix results together to get a better final prediction. we have calculated two confusion matrixs as follows from naive bayes and the decision tree True positive totals and lessen the False negatives.
Topic data-science-model python machine-learning
Category Data Science
There's no way to combine two models from their confusion matrices. The only thing which can be done at this level is to select the best model as the final model.
Combining models is possible at the level of individual predictions. This is called ensemble learning. The most basic method is to assign the majority label using an odd number of models. For example if three models predict 1, 0 and 1 then we select 1 because this answer has more "votes". There are also more complex methods.
Note that the classes in your data are not balanced, class 1 has 35% of the instances. It would be better to use precision/recall/f-score rather than accuracy as a performance measure.
Geeks Mental is a community that publishes articles and tutorials about Web, Android, Data Science, new techniques and Linux security.