Unsupervised Learning - Kmeans Week 11 Challenge
The Car Dataset has been successfully cleaned and encoded, making it ready for K-Means clustering.
Next Steps:
- Normalize numerical features to ensure proper clustering.
- Apply K-Means clustering with an optimal number of clusters.
- Analyze business insights from the clusters.
The Elbow Method plot shows that the optimal number of clusters is around 3 or 4, where the inertia (WCSS) starts leveling off.
Next Steps:
- Apply K-Means Clustering with the optimal number of clusters.
- Analyze the clusters and derive business insights.
The Car Dataset has been successfully clustered into 3 groups using K-Means.
Business Insights from Clustering:
- Cluster 0: Likely contains high-performance or luxury cars with higher engine size, horsepower, and price.
- Cluster 1: Possibly compact or fuel-efficient cars, characterized by smaller engine size, lower horsepower, and better MPG (fuel economy).
- Cluster 2: May include mid-range cars, balancing price, engine power, and fuel efficiency.
Next Steps:
- Visualizing the Clusters to validate these insights.
- Interpreting trends for strategic pricing, marketing, or product positioning.
Business Insights from Clustering:
-
First Plot (Horsepower vs. MPG City)
- Cluster 0 (Dark Purple): Likely represents mid-range cars with moderate horsepower and fuel efficiency.
- Cluster 1 (Teal Blue): Likely compact and fuel-efficient cars, characterized by lower horsepower and high MPG.
- Cluster 2 (Yellow): Likely high-performance/luxury cars, with high horsepower and lower MPG.
-
Second Plot (Weight vs. Engine Size)
- Cluster 0 (Blue): Smaller and lighter vehicles, likely budget-friendly compact cars.
- Cluster 1 (Gray): Cars with moderate weight and engine size.
- Cluster 2 (Red): Heavy and powerful cars, likely SUVs or luxury sedans.
These insights can help:
- Car manufacturers optimize pricing, marketing, and production strategies.
- Dealerships target customers based on fuel efficiency, performance needs, or luxury features.