Latest posts tagged with #featureSelection on Bluesky
Adaptive Feature Selection Enhances Graph Neural Networks
Adaptive node feature selection lets GNNs prune attributes during training, keeping only the most informative ones; tests on GCN and GAT dropped many features without hurting performance. Read more: getnews.me/adaptive-feature-selecti... #gnn #featureselection
BoMGene hybrid feature selection boosts gene expression classification
BoMGene merges Boruta and mRMR, evaluated on 25 gene‑expression datasets, cutting feature count while keeping or improving classification accuracy. getnews.me/bomgene-hybrid-feature-s... #bomgene #featureselection
Random Feature Subsets Challenge Feature Selection in High‑Dimensional Data
Random subsets of 0.02‑1 % of variables matched or outperformed full and selected feature sets in 28 of 30 high‑dimensional datasets across RNA‑Seq and imaging. Read more: getnews.me/random-feature-subsets-c... #featureselection #randomsubsets
Nonconvex Regularization Boosts Feature Selection in RL
Batch algorithm uses a PMC penalty and FRBS optimization to prune RL features, dropping irrelevant variables while preserving policy performance on benchmarks. Read more: getnews.me/nonconvex-regularization... #reinforcementlearning #featureselection #nonconvex
Can we use statistical tests to select features? 🤔
Turns out, we can! 🎉
In the slides below, we’ll explore the most commonly used statistical tests for feature selection, along with their advantages and limitations. 👇
#machinelearning #datascience #featureselection
Feature selection is key to robust Machine Learning models.
Feature-engine helps you pick the best features, from removing duplicates to advanced methods like permutation.
Watch how to optimize your models here!
📹https://f.mtr.cool/ljhsellcdv
#MachineLearning #featureselection
📘 Feature Selection in Machine Learning – 2nd Edition
The only book fully dedicated to feature selection.
Learn how to choose the right features for simpler, faster, and more interpretable models.
👉 Our book: https://f.mtr.cool/ydrvbttjdh
#FeatureSelection #MachineLearning
Identifying Key Predictors of Smoking Cessation Success: Text-Based Feature Selection Using a Large Language Model
Hu, Y., Le, T. T. T. et al.
Paper
Details
#SmokingCessationSuccess #FeatureSelection #LargeLanguageModel
A Hybrid Deep Learning–Based Feature Selection Approach for Supporting Early Detection of Long-Term Behavioral Outcomes in Survivors of Cancer: Cross-Sectional Study #CancerSurvivors #DeepLearning #FeatureSelection #MachineLearning #HealthTech
When training #machinelearning models, which features actually matter? 🧠
A quick way to find out is by training single-feature models.
🎥 Watch: https://tinyurl.com/yc7aje5r
📚 Learn more: www.trainindata.com/p/feature-selection-for-...
#featureselection
Spending hours researching the best practices for feature selection? ⏲️
If so, watch this video 📹 to see how this course can help you build simpler, faster, and more robust machine learning models. 👇
📹https://tinyurl.com/n6fyd6ry
#machinelearning #featureselection #datascience
Beyond filter, wrapper and embedded methods, there is a whole world of feature selection algorithms.
Good news is... most of them are available in Feature-engine.
You can find the link in the comment below 👇
https://tinyurl.com/288yhd7d
#featureselection #featureengine
With Feature-engine’s SelectByTargetMeanPerformance, you can:
✅ Encode categorical variables
✅ Transform numerical features
✅ Select the most predictive ones
Watch the video and learn how to apply this technique to your ML models!👇
https://tinyurl.com/2w5ntwyz
#featureselection #machinelearning
A strong feature subset is highly correlated with the target but not with each other.
This thesis dives deep into correlation-based feature selection and likely inspired methods like MRMR.
Want to learn more? Check it out!👇
📄https://tinyurl.com/2nhjr3x2
#featureselection #machinelearning #ml
One way to rank variables is by testing a classifier or regression model on each one individually. The image shows how it works!
Want to dive deeper? Check out my book for a detailed breakdown with step-by-step procedures with illustrations. 📖✨
https://tinyurl.com/yc6sxvvu
#featureselection
Watch how Recursive Feature Addition with Feature-engine optimize feature selection! We build a classifier, derive feature importance & select the best features—1 at a time—using cross-validation for better accuracy. 📊✨
📹https://tinyurl.com/mrxskrr2
#MachineLearning #FeatureSelection #ml
Can we use statistical tests to select features? 🤔
Turns out, we can! 🎉
In the slides below, we’ll explore the most commonly used statistical tests for feature selection—along with their advantages and limitations. 👇
More resources - my book 📘 https://tinyurl.com/yc6sxvvu
#featureselection
Picking the right features is key to building a strong, interpretable #ML model. #Featureselection helps cut through the noise, improving accuracy and resilience.
In this blog, you’ll learn filter-based selection methods, how they work, and when to use them.👇
https://tinyurl.com/2a2u7xa4
Beyond filter, wrapper and embedded methods, there is a whole world of feature selection algorithms.
Good news is... most of them are available in Feature-engine. 👍
https://buff.ly/45uNPAW
#featureselection #featureengine #machinelearning #datascience #dataengineering #mlmodels #ml #ai #algorithms
#Featureselection with Feature-engine made easy! 😊
I’ve put together a playlist so you can learn to:
✨ Drop constant & quasi-constant features
✨ Remove duplicates with Feature-engine & Scikit-learn
✨ Pick the most predictive features and more...
Watch now 👉📹 https://tinyurl.com/ywppy8j2
You can now implement MRMR out of the box with Feature-engine.
More details about how the method works in Feature-engine's documentation.👇
https://bit.ly/3ULpao9
#featureselection #datascience #dataengineering #AI #machinelearning #ML
The "HMDA" #R #Package is finally out. This package implements #Holistic #Multimodel #Domain #Analysis, a new #machinelearning paradigm for stabilizing the result of #exploratory #analysis, performing #automatic #featureSelection, and finding important #factors and #domains in #bigdata.
More features aren’t always better! While adding variables won’t necessarily hurt accuracy, using fewer, well-chosen features comes with big benefits. 🙂
Want to dive deeper? Check out the details here:👇
www.trainindata.com/p/feature-selection-in-m...
#featureselection #ml
This is the go to review to know more about feature selection for machine learning.
I am yet to find a better article than this one, even though it is from 2003.
If you are new to feature selection, this is your starting point. 👍
https://dl.acm.org/doi/pdf/10.5555/944919.944968
#featureselection
🚀#Featureselection made easy!
In this video, we show how to use Recursive Feature Addition with Feature-engine to find the best features for your model.
Watch as we build a classifier, assess feature importance, and improve performance—one feature at a time!👇
https://bit.ly/4kmYLIA
A Hybrid Deep Learning–Based Feature Selection Approach for Supporting Early Detection of Long-Term Behavioral Outcomes in Survivors of Cancer: Cross-Sectional Study #CancerSurvivors #DeepLearning #FeatureSelection #MachineLearning #HealthCare
New to feature selection? Want to learn what methods are out there?
Then check out one of the iconic reviews on feature selection for machine learning.
A great way to get an overview, for free. 👇
https://tinyurl.com/3hv8fwjd
#featureselection #machinelearning #datascience #dataengineering
L1 vs L2 for feature selection—what’s the difference? 🤔
L1 and L2 regularization serve unique purposes in feature selection. Want to know why?
Check out the slides to find out!
#MachineLearning #DataScience #FeatureSelection #L1Regularization #L2Regularization #AI #ML
We identify the relative importance of the variables, even when they have different units of measures, and produce descriptions which are at the same time 𝐥𝐨𝐰-𝐝𝐢𝐦𝐞𝐧𝐬𝐢𝐨𝐧𝐚𝐥 𝐚𝐧𝐝 𝐢𝐧𝐭𝐞𝐫𝐩𝐫𝐞𝐭𝐚𝐛𝐢𝐥𝐞. #featureselection
PS: 📅 #HELPLINE. Want to discuss your article? Need help structuring your story? Make a date with the editors of Low Code for Data Science via Calendly → calendly.com/low-code-blo...
#datascience #featurereduction #ml #weka #featureselection #KNIME #lowcode #nocode #opensource #visualprogramming
