Questions tagged [feature-selection]

Methods and principles of selecting a subset of attributes for use in further modelling

Filter by
Sorted by
Tagged with
6 votes
4 answers
598 views

Can the χ² test be used without a contingency table?

I thought that the chi squared (χ²) test is to be used when one has an r × c contingency matrix, i.e., when the dependent variables are nonnegative, span the same <...
  • 645
0 votes
0 answers
21 views

What is the best model selection method for high-dimensional linear regression?

Model selection (best subset selection) in linear regression is quite important in many applications. Among the methods belonging to different frameworks such as information criterion, hypothesis ...
0 votes
0 answers
10 views

Knockoff filters simple explanation and importance!

Knockoff filters are new in the field of variable selection. Can someone provide (or refer to) a slightly simple understanding of the topic? Also, what is the fuzz about this new method compared to ...
0 votes
1 answer
29 views

What are some good feature selection techniques for binary classification problems?

I am performing binary classification on a data set with around 90k records and 28 features. I'd like to evaluate various models such logistics, SVM, Xgboost etc. via grid-search method to see which ...
  • 11
0 votes
0 answers
25 views

Are my samples not independent?

I want to predict soccer games and want to calculate the statistical significance of my features for selection. One of my features is the amount of points both teams have before the game. Does this ...
  • 13
0 votes
1 answer
22 views

Principle Component Analysis for Feature extraction from Voltage and Current Signals

I am doing research work on fault classification in power transmission lines. I generated fault datasets in MATLAB/Simulink and collected it in matrix format with 6 various i.e. 6 variables in 6 ...
1 vote
0 answers
19 views

Random forests in low-dimensional vs high-dimensional datasets

Summary: Is random forest naturally more robust to bad features in high vs low dimensional datasets? I've made a number of text classifiers with Scikit's Random Forest, with thousands of features from ...
0 votes
0 answers
10 views

How can I compare two vectors of fitted values?

I'm testing out an algorithm for the lasso and it produces fitted values. I want to compare these to the fitted values obtained from glmnet. I thought the Euclidean ...
  • 151
2 votes
0 answers
29 views

Does selecting confounder variables for a model with multiple correlation tests risk biasing results (similar to forward selection)?

My team is conducting a counterfactual difference-in-differences (DiD) healthcare analysis to estimate the benefits of home nursing visits compared with a control group. We've "pre-selected" ...
  • 4,706
1 vote
0 answers
10 views

How to engineer and test seasonality features?

My task is multiclass classification of item to buy (next). I have a purchase history dataset with a datetime feature. From it I could engineer many new seasonality features: Time of year (season, ...
0 votes
0 answers
10 views

How to extract features of an unseen test image

Initially there are some images(medical images). We have designed couple of feature extraction techniques. Now lets say we are getting 10000 features for a particular image, its not really good to use ...
0 votes
0 answers
13 views

How to select the best binary inputs for a Bayesian model?

I am generating a binary Bayesian model to predict if I am asleep. I set the Prior P(Asleep) based on historical data (8 hours a day = 0.333), and for arguments sake I am using a posterior threshold ...
  • 131
1 vote
1 answer
44 views

role of onehotencoding in Feature selection

I need to clarify about feature selection. I am working on Kaggle breast cancer dataset (https://www.kaggle.com/datasets/reihanenamdari/breast-cancer). It is a categorical dataset. There are 15 ...
  • 133
0 votes
0 answers
16 views

Pipeline design for creating regression/classification models of brain connectivity matrices to predict clinical improvement

I've been tasked with using an EEG based brain connectivity dataset(26 stroke subjects) to predict clinical recovery. The connectivity matrix for each subject consists of connectivity values for 190 ...
1 vote
0 answers
34 views

Total contribution of feature using SHAP values in multi-output regression problem

I am interested in using SHAP values to perform wrapper-based feature selection. The model has 3 outputs, and I do have SHAP values for each output, using GradientExplainer, but even after a bit of ...
0 votes
1 answer
40 views

Choosing parameters for an artificial neural network with a likelihood ratio test

I am currently trying to choose which parameters to use in my artificial neural network. Because the end goal is a comparison between the logistic regression and the neural network and I have already ...
  • 43
0 votes
0 answers
14 views

error metrics resulted from time series forecasting as feature

While reading a paper about Time Series Anomaly Detection (https://www.usenix.org/conference/atc19/presentation/zhang-xu) I came to this : "Forecasting error features: Following the prior work [...
0 votes
0 answers
19 views

Feature engineering in frequency-domain time series

I am solving a task of a real-time binary signal (electric current) classification. This is a project that I am continuing so I have been supplied with a feature extraction part already done. The ...
  • 1
1 vote
0 answers
25 views

Choosing the correct set of covariates (i.e. confounders) for inverse probability weighting

I use inverse probability weighting (IPW) to estimate the impact of a marketing intervention in the retail industry. There is a test group of stores and a control group of stores, and the ...
0 votes
1 answer
35 views

Should I correct for batch effect before selecting features using random forest for RNA-Seq data?

This is a mix of bioinformatics and ML problem. Hope someone with both expertise can help. Please forgive me if it's unclear or I used the wrong words as I am very new to ML. I am trying to pick out ...
  • 1
1 vote
0 answers
23 views

How do you set a "score" of the data? [closed]

My data is like this ...
1 vote
0 answers
37 views

Can we use Standard Deviation for feature selection?

I am working on the House Price Prediction dataset on Kaggle and am trying to identify the good features for our price predictions. For numerical variables, I have gone with a high correlation with ...
  • 111
0 votes
0 answers
26 views

OLS After LASSO to remove negative coefficients

I have a regression model with many predictors and not that many instances. (~70 predictors and 150 instances) I would like to use the model for inference, and therefore need to identify the sparse ...
  • 203
0 votes
0 answers
8 views

Feature selection based on production data

I have a classifer (one/zero labels) that was trained and hypertuned by the book. When the model was ready, I applied it to the production data: real-time and unlabeled. After a short period (a few ...
  • 77
0 votes
1 answer
43 views

Feature selection with categorical variables

I am working on the diamonds dataset. In it, we model the price of those diamonds based on several predictors. Three of them are categorical variables (cut, color and clarity) and the rest are ...
  • 111
0 votes
0 answers
52 views

Computing confidence intervals backward elimination with bootstrapping

I do backward elimination, by iteratively removing the biggest p-values until the biggest p-value is < 0.157. Then, I have a model, which confidence intervals displayed are not wide enough: "...
2 votes
1 answer
46 views

Methods to identify the optimal number of bins between two groups

I am training a supervised machine learning model. The training data contains 2 independent groups of people. The dataset contains independent continuous variables and 1 dependent binary variable. I ...
  • 23
0 votes
0 answers
24 views

Understanding and combining feature selection and cross validation for random forest

Suppose we are interested in the random forest classifer and the hyperparameters are n_estimators:[100,500,1000], max_depth: [1,5,15] This gives rise to 9 ...
  • 1,018
1 vote
0 answers
32 views

Classifying dataset with different number of features

I have a dataset like below: samplename position reference alternative S1 201 C T S1 3567 A G S1 760 T C S2 356 C T S2 6787 T C These data belongs to patients and ...
  • 11
1 vote
1 answer
37 views

Feature subset selection decision rule

I want to select a feature subset. I know that every feature in this dataset is informative to some level, due to domain knowlege. So in theorey I should use all features to maximize the negative mse ...
  • 43
0 votes
0 answers
17 views

Dealing with interaction effects between fractional variables ... Does it matter that they're fractional?

I have a number of variables, $X_1, X_2, ...$ which vary between 0 and 1. These are measures representing metrics in a large organization -- e.g: LeadershipScore on a scale of 0 to 1, ...
  • 219
1 vote
1 answer
66 views

Features differ between classes

Good evening everyone. Regarding the topic related to Sparse Clustering (for example K-Means). For example, in "Witten DM, Tibshirani R. A framework for feature selection in clustering" the ...
2 votes
1 answer
567 views

Can a variable be linearly independent, but non-linearly dependent?

I am reviewing a friend's paper, and they are throwing out variables that are below a certain correlation coefficient value before doing a multiple linear regression model. Is this a wise thing to do? ...
1 vote
0 answers
17 views

Multiple Context Time-Series Variable Selection

I'm working with a time-series data set where I'm given an entities', quarterly, total balance (measured in billions of dollars) over 15 years (dependent variable). I'm also given 16 explanatory ...
1 vote
1 answer
11 views

Lowering the weight of particular features in a neural network?

Given sample data $x$, we hypothesize that some features (i.e. dimensions) of $x$ will generalize well, while others will generalize poorly. For example, when predicting medical diagnosis, age and ...
1 vote
0 answers
37 views

Feature selection via RFE, MRMR, embeded methods and categorical features' impact

I am using ensemble-tree for regression (in Matlab) for my research. I have 22 features that includes 16 continuous (numerical) and 6 categorical variables. Categorical variables are based on time, ...
  • 11
0 votes
0 answers
34 views

feature selection within large dataset

I have a dataset containing more than 1000 predictors and I would like to do the feature selection. The features belong to several big categories(geographical factors, customer information....etc) ...
0 votes
1 answer
27 views

Hyperparameter tunning in SelectKBest feature selector

I am working with a pretty large dataset containing 760 rows and arround 58k-60k features and I'd like to perform a feature selection to reduce the dimensionality of those. After stardardising the ...
  • 23
0 votes
0 answers
13 views

Can I select features which are a split-up of the target variable for a regression model?

I'm currently working on predicting the energy consumption of a house (in KWH). After preprocessing the dataset, I used XGBRegressor to find the importance of features (feature importance method). The ...
3 votes
2 answers
153 views

Is this a correct interpretation of percent importance?

In this scenario, the end goal is to determine which columns are most important. I use this term very very very loosely, because I know that interpretability is never straightforward but I just want ...
2 votes
1 answer
43 views

Which data preprocessing steps do I need to perform on which data subset?

I want to make predictions using several supervised Machine Learning algorithms and apply 10-fold-cross validation. For doing so, I randomly divided my dataset into in-sample and out-of-sample sets. ...
3 votes
1 answer
110 views

Improving a forest model by dropping features below a percent importance threshold?

I'm wondering if there is a term for this process, where I can find more reading/information about it, and if this is valid or hacky overall. I've only used this process for tree-based models which ...
2 votes
1 answer
35 views

Can Correlation based feature selection discard features that show no correlation by themselves but are meaningful only if combined?

Assuming a feature selection process based on correlation or some other metric, is it possible to overlook input features that by themselves show no actual correlation with the target values, but that ...
19 votes
3 answers
2k views

Widespread overfitting in health domain research?

I was reading about flaws with model selection techniques such as elimination based on significance and backwards selection via AIC (or similar) in the context of regression leading to inflated ...
  • 335
1 vote
0 answers
33 views

Which kind of normalizations is better?

I want to do feature selection for a linear model. For this, each feature coloumn should be normalized, and a useful manner is substracting the mean and divided the l2 norm (to obtain unit norm): $\...
  • 11
0 votes
1 answer
35 views

Selecting features for classification model

I have a dataset that contains some numeric variables. What im doing is trying to check how important these variables are in describing endogenous variable in order to include them in my classifier. I ...
2 votes
1 answer
106 views

sklearn's permutation_importance returns surprising result

I have simulated normally distributed data (x_1 = np.random.normal(0, 1, size=1000)) and used it to create a dependent variable with a linear combination of the ...
  • 63
1 vote
0 answers
25 views

How does the weight work in gglasso R package

I am using gglasso package to perform regularized weighted least square for grouped variables. But I was confused with their ...
1 vote
0 answers
24 views

Undersampling and Oversampling

I've an unbalanced dataset. I need do to perform feature selection and then I'm going to fit my model. Is it conceptually wrong doing undersampling and perform feature selection and then once ...
0 votes
0 answers
27 views

Feature elimination to screen for multiple models using tidymodels

I am currently performing regression modeling, with a dataset that has number of features (p) higher than observations (n). Typically p = 10000 and ...

1
2 3 4 5
46