Covariance Matrix in Mean-Variance Optimization

• Points that have the highest average path length
• Points with the highest density of neighbors
• Points with a shorter average path length than normal points (correct)
• Points with the longest distance from the centroid

• Group closely packed points together (correct)
• Identify isolated points in high-density regions
• Mark all points as outliers
• Group points randomly to form clusters

• To speed up the algorithm
• To prevent overfitting and increase robustness (correct)
• To guarantee all outliers are identified
• To ensure all points are considered equally