How to build machine learning models for imbalanced datasets

preview_player
Показать описание
In this video, we will explore the important topic of how we can build machine learning models for imbalanced datasets. Particularly, for classification problem where the classes are imbalanced we can perform class balancing (either via undersampling or oversampling). This video benchmarks or compares various class balancing approaches along with the imbalanced dataset as a control for model building.

🌟 Join as a Channel Member to support us:

⭕ Watch this video next:

⭕ Support my work:

⭕ Recommended Books:

⭕ Disclaimer:
Recommended books and tools are affiliate links that gives me a portion of sales at no cost to you, which will contribute to the improvement of this channel's contents.

⭕ Stock photos, graphics and videos used on this channel:

#datascience #machinelearning #dataprofessor
Рекомендации по теме
Комментарии
Автор

🌟 Join as a Channel Member to support us:

DataProfessor
Автор

Thank you so much for this Professor :)

Pedro_Gorla
Автор

Yo Data Prof, thoughts on using SMOTE? Was building some stuff this morning and wondered how the pros (aka you 😉) are fitting it into pipelines. I figured you'd have to run the oversampling before any feature engineering otherwise the features may not be as accurately generated as opposed to doing it by hand. comments in your code 🙌🙌🙌 soothing as hell for my OCD.

NicholasRenotte
Автор

Hello. I have a question. So with respect to building the model without balancing the class i observed that you used cross-validate from scikit-learn. WIll gridsearch cv be able to do the same job ?

edmundquarm
Автор

Very useful tutorial
Can you do a video about *SMOTE* and *ktrain package*
Because until now i am not abel to make it work.
Thank you in advance.

lebdev
Автор

good job, how about multiclass datasets, thank you so much

imadful
Автор

sir you give ideas of machine learning to start bioinformatics base project thatturnsn to publication

MuhammadFaizan-miyo
Автор

Just a question, is it better to split the train and test set using stratify option? Before we apply balancing to it? Since we are dealing with imbalanced data. I think it would help the to get better score for the test set

mexnibip
Автор

keep doing the good work, as I say always (I enjoy your videos)

adnaneaouidate
Автор

Thanks a lot Data Professor. Your videos have been helpful in executing my Bioinformatics projects.

sulaimonridwan
Автор

What’s the probability of YOUR job being replaced by a robot?
We want to help you stay informed with our bite sized data!

DragonData