edit README

Author: alexattia

KaggleTaxiTrip/README.md

@@ -7,13 +7,17 @@ The first part is to analyze the dataframe and observe correlation between varia
 ![Distributions](https://github.com/alexattia/Data-Science-Projects/blob/master/KaggleTaxiTrip/pic/download.png)
 ![Rush Hour](https://github.com/alexattia/Data-Science-Projects/blob/master/KaggleTaxiTrip/pic/rush_hour.png)
 
-### Second part - Cleaning and feature selection 
+### Second part - Clustering
+The goal of this playground is to predict the trip duration of test set. We know that some neighborhoods are more congested. So, I used K-Means to compute geo-clusters for pickup and drop off.
+![Cluster](https://github.com/alexattia/Data-Science-Projects/blob/master/KaggleTaxiTrip/pic/nyc_clusters.png)
+
+### Third part - Cleaning and feature selection 
 I have found some odd long trips : one day trip with a mean spead < 1km/h. 
 ![Outliners](https://github.com/alexattia/Data-Science-Projects/blob/master/KaggleTaxiTrip/pic/outliners.png)
 I have removed these outliners. 
 
-I also added two features from the data available : Haversine distance and Manhattan distance.
+I also added features from the data available : Haversine distance, Manhattan distance, means for clusters, PCA for rotation.
 
-### Third part - Prediction
-I am currently using a Random Forest. 
-Current Root Mean Squared Logarithmic error : 0.45
+### Forth part - Prediction
+I compared Random Forest and XGBoost. 
+Current Root Mean Squared Logarithmic error : 0.391