Missing values constituted less than 2% of all data points and were addressed using the expectation–maximization algorithm. Multivariate outliers were identified using Mahalanobis distance (critical ...

Algorithms and Data Structures - 2nd Workshop, WADS '91 Ottawa, Canada, August 14–16, 1991 Proceedings, Frank Dehne & Jörg-Rüdiger Sack (1991) [pdf] Uncertainty in Knowledge Bases - 3rd International ...

Missing values were imputed using the Expectation–Maximization method for continuous variables and the mode for categorical variables. Patients with more than 20% missing data were excluded.

Customer stories Events & webinars Ebooks & reports Business insights GitHub Skills ...