Decision Tree

Now we will look at the results of all the experiments we did for predictive data mining.

Before Tuning

Experiment 1: Resampling only

60:40 Ratio

From the result below, Python has higher accuracy than RapidMiner which is 62.35%. As we can see from the results, Python was able to predict the health status more precisely than RapidMiner. Hence, for this ratio, Python is a better model than RapidMiner.

80:20 Ratio

Based on the two results, for 80:20 ratio, Python is still the best model with an accuracy of 63.79% while RapidMiner only has an accuracy of 55%. For both ratios, Python is considered as the best model with higher accuracy and able to predict the health status precisely.

Experiment 2: Resampling and Standardization

60:40 Ratio

From the result below, Python has slightly higher performance accuracy in which Python is a better model for health status prediction compared to RapidMiner. The prediction for "Not Healthy" is lower compared to Python which means the RapidMiner got more wrong predictions than Python.

80:20 Ratio

For 80:20 ratio however, RapidMiner has higher performance accuracy so RapidMiner was able to predict more precisely than Deciosn Tree using Python.

Experiment 3: Resampling, Standardization and Discretization

60:40 Ratio

From the result below, Python was able to predict the financial level more precisely than RapidMiner with a classification error of 37.65%.

80:20 Ratio

For 80:20 ratio, RapidMiner is the best model since the prediction is more accurate then Python. RapidMiner has an accuracy of 59.09% while Python has an accuracy of 56.38%.

After Tuning

Experiment 1: Resampling only

60:40 Ratio

From the result below, the accuracy for RapidMiner is higher than Python which is 63.37%. RapidMiner was able to predict the financial level more precisely than Python.

80:20 Ratio

For 80:20 ratio, RapidMiner still has higher performance accuracy than Python which is 64.29%. Just like 60:40 ratio, RapidMiner was able to predict the financial level more precisely than Python. So for this experiment, we can conclude that RapidMiner is the best model.

Experiment 2: Resampling and Standardization

60:40 Ratio

From the result below, the accuracy for RapidMiner is way higher than Python which is 65.02%. RapidMiner was able to predict the financial level more precisely than Python.

80:20 Ratio

From the result below, the accuracy for RapidMiner is higher than Python which is 67.36%. Thus, RapidMiner was able to predict the financial level more precisely than Python. For this experiment, RapidMiner is still the best model just like experiment 1.

Experiment 3: Resampling, Standardization and Discretization

60:40 Ratio

From the result below, the accuracy for RapidMiner is higher than Python which is 64.40%. RapidMiner was able to predict the financial level more precisely than Python.

80:20 Ratio

From the result below, the accuracy for RapidMiner is way higher than Python which is 67.36%. RapidMiner is still the best decision tree odel to use for this prediction.

To sum it all up, some results are way higher before hyperparameter tuning and some are not. It actually did not meet our expectation as we were expecting the result to increase but it turned out differently. As for this model decision tree, the overall results are better and higher after hyperparameter tuning compared to before.