By using RapidMiner it shows that Random Tree is better than Decision Tree because the accuracy results are more stable. Especially when the bin value is changed, Random Forest result is still high compared to Decision Tree whose value drop sharply. This is likely because Random Forest has a higher training time than a Decision Tree which will make it more accurate than Decision Tree.
Similar to Google Colab, the result of Random Forest shows the better performance as the accuracy increases steadily while on the Decision Tree, there is a decrease when tuning.
In both application shows that Random Forest is more stable considering to the theory that Random Forest will provide many beneficial results as it gather a lot of results to limit overfitting as well as error due to bias. Thus, make Random Forest a powerful model.
Last but not least, by using different method for K-Means Clustering such as Elbow method and Silhouette Score, we can find the optimal number for clustering. Both of the method is really helpful in order to fine the optimal number of cluster, however by using Elbow Method it is more simple and faster.
Through data analysis we can know and predict various information. By using this information, we can improve many things such as from the basketball analysis itself will help the stakeholder such as Company Organization to review the performance of the player and give sponsorship to the player who are benefits for them and promote PelontarXI team based on the player profile. They can highlight the best performance for each player. As for the weak players, the coach as a stakeholder can monitor their performance and give them special training based on the player's weakness. Indirectly, this will also help to balance the player's strength so that the player will have no problem in changing for any position either Offensive or Defensive.
Becoming proficient in using Rapidminer and coding in Google Colab is a very valuable skill because it is able to hone the talent to continue to progress in this field. In addition, both of Decision Tree and Random Forest have their pros and cons depending on how we use them and set them up.