In this section, I present the next part of the analysis regarding the segmentation of the real estate dataset.

https://github.com/PawelTokarski95/Real-Estate-Segmentation

​

I created a total of three models by importing the appropriate libraries: Decision Tree, SVM, and Random Forest. For all these classifiers, I generated detailed classification reports, including metrics such as recall, accuracy, and precision, which are essential for evaluating model performance. Additionally, I used cross-validation to ensure the validity and reliability of the models' functioning.

​

Decision Tree classifier:

​

​

​

​

​

​​

​

​

​

​

​​​​​​

​​

 

 

 

 

 

 

 

SVM classifier:​

​

​

​

​

​

​

​

​

​

​

​

​​​​

​​

​

​

​

​

​​​​​​​

​

​

​​​​​​Random Forest classifier:

​​​​

​

​

​

​​

​

​

​

​

​

​

​

​

​​​​​​​​​​​​​​​

​

​​​​​​

 

 

 

Cross-validation scores:

Decision Tree classifier:

0.70

SVM classifier:​

​0.90

Random Forest Classifier:

0.80

​

Based on the attached model evaluations, it all models achieved very good results and are pretty reliable in terms of cross-validation score. This indicates that I would select that particular models for predictive purposes.

​