Isolation Forest: A Tree based approach for Outlier Detection (Clearly Explained)

[@roi_4_dayz]

Very nice explanation. Thanks for the help!

[@aadhyatiwari9688]

One of the best explanations out here! Thankyou

[@HafeezUllah]

Very nice explanation.

[@ankithahj1694]

Very good explanation. Suppose for single set a feature values, I am getting as outlier. How to find out top features contributed for the decision?

[@prathipmathavan3089]

helped me a lot, keep up he good work!

[@VikasVerma-xf6hb]

Nice ...Thanks

[@azogdevil]

Thanks 🙏

[@marvaajab6657]

i have face issue to get the high performance of this model this model are total 9 parameters but hyperparameters are which one?

[@machinelearningplus]

Want to learn more ML? Checkout edu.machinelearningplus.com/s/pages/ds-career-path - Become fundamentally strong in Data Science and ML!

[@ashrithadepu]

Hi, my dataset consists of categorical values and I’ve label encoded them to use isolation forest model. But how to evaluate my model? What metrics should I follow?

[@machinelearningplus]

If your 'features' are categorical, don't label encode then. Label encoding is meant for Target variables. Evaluating models can be done as you would with any other predictive model

[@ashrithadepu]

@@machinelearningplus but I don’t have a target variable and all the data is categorical how do you think I can proceed?? Btw thanks for your reply

[@r0cketRacoon]

@@ashrithadepu one hot encoding, that'all

[@mikeclark4611]

🙂 Promo sm