Algiers Apartment Price Prediction
This is a set of regression algorithms from Scikit-Learn applied to a dataset that I created by scraping Ouedkniss.
This dataset has a total of 3553 samples with a size of 13, features are real positive values and targets are real between 1000 and 3000.
Algorithms and results for dataset
Algorithm |
Model performance |
Linear Regression + Polynomial Features | 0.02067345956704003 |
Gradient Boosting Regressor | 0.12610201107980723 |
K-Neighbors Regressor | -0.1956575117534145 |
Decision Tree Regressor + Ada Boost Regressor | 0.025200951666729643 |
Decision Tree Regressor + Bagging Regressor | 0.1218648622524763 |
Random Forest Regressor | 0.1303632423343436 |
Huber Regressor | -0.2150968619259983 |
Theil Sen Regressor | -0.03935666883265565 |
Linear Regression + Shuffle | 0.052494793812623786 |
Correlation of selected features for dataset
|
Commune |
Etage |
Superficie |
Piece |
Prix |
Commune |
1.0 |
-0.15976914085263527 |
0.20950954048367096 |
0.05282241062499728 |
0.0770504304324702 |
Etage |
-0.15976914085263527 |
1.0 |
0.02088774679844994 |
0.03225100204879875 |
0.0028625214924493715 |
Superficie |
0.20950954048367096 |
0.02088774679844994 |
1.0 |
0.7423177491115521 |
-0.01459807463078822 |
Piece |
0.05282241062499728 |
0.03225100204879875 |
0.7423177491115521 |
1.0 |
-0.02782093484314777 |
Prix |
0.0770504304324702 |
0.0028625214924493715 |
-0.01459807463078822 |
-0.02782093484314777 |
1.0 |
Data Profiling for dataset
|
Minimum |
Maximum |
Mode |
Mean |
Median |
Standard deviation |
Quantile [0.25, 0.5, 0.75] |
Commune |
1 |
28 |
4 |
13.17117117117117 |
13.0 |
8.574879777660525 |
[4.0, 13.0, 22.0] |
Etage |
1 |
15 |
1 |
3.1331644144144146 |
3.0 |
2.174971764738508 |
[1.0, 3.0, 4.0] |
Superficie |
1 |
250 |
75 |
95.88006756756756 |
90.0 |
33.34758832480636 |
[75.0, 90.0, 115.0] |
Piece |
1 |
9 |
3 |
3.516328828828829 |
3.0 |
0.8903647264348437 |
[3.0, 3.0, 4.0] |
Prix |
1000.0 |
3000.0 |
2000.0 |
1905.3470157657657 |
1800.0 |
553.3634995790693 |
[1450.0, 1800.0, 2350.0] |
Plot features for dataset
![commune](https://user-images.githubusercontent.com/19336306/42030378-87190468-7aca-11e8-8c64-60f721b4ce4d.png)
![etage](https://user-images.githubusercontent.com/19336306/42030380-87579d9a-7aca-11e8-86a0-97f9606e0bac.png)
![piece](https://user-images.githubusercontent.com/19336306/42030381-8794c422-7aca-11e8-9c08-7250624f36b8.png)
![superficie](https://user-images.githubusercontent.com/19336306/42030382-87c79118-7aca-11e8-8141-4a55219ecfaf.png)
Python 2.7 and up
The followoing are the prerequiste Python modules that needs to be installed to execute main.py:
sudo pip install pandas
sudo pip install -U scikit-learn
sudo pip install numpy
sudo pip install matplotlib
Clone the repository using the below mentioned command and execute the Python program.
git clone https://github.com/Sehaba95/Algiers-Apartment-Price-Prediction.git
cd Algiers-Apartment-Price-Prediction/Regressors
python AdaBoostRegressor.py
Okba BEKHELIFI