I am identifying outliers using K-means and LOF (Local Outlier Factor). Let's say if we are identifying possible outliers using both the techniques, I believe LOF will pick global outliers also as they will also be local to any cluster . LOF distance of outliers identified using K-means will always be high as they will be far away from n neighbors.

Although value of n ( neighbors) will affect LOF outliers but in broader term can I say K-means identified outliers will be subset of LOF outliers??

K-means people know well.Easy article on LOF is https://en.wikipedia.org/wiki/Local_outlier_factor

My data has multiple operating regions so I am planning to go with LOF but I also believe it should incorporate most of the K-means outliers. Please guide me through this!



Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.