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

learn more… | top users | synonyms (2)

0
votes
2answers
80 views

Selecting features and estimating their out-of-sample performance with cross-validation

I have only a small dataset. I want to 1. select the most predictive features out of a large candidate pool and 2. get an estimate of their expected predictive performance. In the elements of ...
2
votes
0answers
100 views

Quantify the information lost given by the Kullback-Leibler divergence measure

Consider there are $N$ individuals and these measure a quantity $X\in \mathbb{R}^{N\times M}$ where $M$ is the number of measurements and let $P(X)$ denote a probability distribution over $X$. The ...
2
votes
1answer
38 views

Term selection in a model

In selecting terms to include in a model, say a linear one, should we always test the significance of the main effects first, keep only the significant ones, then consider the possible interactions ...
4
votes
2answers
798 views

Cross-validation and feature selection of a multivariate regression

I've been trying to create a multivariate regression model to fit my training data into the prediction of a value. I've put my data into a matrix X with ...
2
votes
0answers
472 views

Linear regression for feature selection

Imagine we regress y on x1...x4. Now, we want to find out if ...
5
votes
3answers
272 views

Regression in $p\gg N$ setting (predicting drug efficiency from gene expression with 30k predictors and ~30 samples)

I have a dataset of 29 cell lines and the IC50 values of a test drug. I want to find a relation between the gene expression profiles of each cell line (nearly 31000 genes) and the IC50 values. My ...
4
votes
1answer
920 views

Interpreting the lasso coefficients

I have used lasso logistic regression on some data and I have some non zero coefficients for some of the features. I want to know based upon the coefficient values how do I rank the features?
1
vote
1answer
61 views

How to compare features and classifiers which achieve perfect accuracy?

So I'm looking to compare different combinations of features and classifiers. But I'm getting a lot of combinations that achieve 100% cross validation accuracy. I'm trying to figure out how I would ...
4
votes
3answers
115 views

How to compare features and classifiers which achieve perfect accuracy?

So I'm looking to compare different combinations of features and classifiers. But I'm getting a lot of combinations that achieve 100% cross validation accuracy. I'm trying to figure out how I would ...
2
votes
1answer
72 views

Feature selection while retaining a specified feature

Pardon if this question is very basic, but I am not able to find any solution for my problem. I am trying to run a feature selection scheme on N features for my classification model, however I want ...
4
votes
1answer
385 views

Variable selection for regression - the subselect package

No regular here will be unaware of the perils of using stepwise and similar automatic methods for variable selection in regression analysis. But preferred alternatives, such as the lasso or ...
1
vote
0answers
160 views

Music classification into genres

I need to classify music (songs) into genres (rock, french house, trash metal, etc). My idea was to extract features from the songs (bmp, zero crossing, etc) and then apply known classification ...
0
votes
1answer
276 views

Recursive feature elimination with only two classes

Recursive feature elimination (RFE) is a feature-selection strategy. It performs in two nested levels of cross-validation. First it tries to divide the training set into N folds. RFE puts one fold ...
1
vote
0answers
180 views

Classifier predicts only one class

I was trying myself in kaggle CIFAR competition, I trained lots of classifiers but get the same result/fail (don't know how to treat them), maybe someone could help me figure what i'm doing wrong. ...
3
votes
1answer
4k views

Understanding the output of C5.0 classification model using the CARET package

The C5.0 classification model was used in this 4-class problem data with $N_{train}$=165, $P$=11, using caret R-package by ...
1
vote
0answers
34 views

Estimating confidence of a prediction

Given a set of features vectors $X=\{\vec{x}_1,..,\vec{x}_n\}$, binary ground truth data $Y=\{y_1,..,y_n\}$ and continuous prediction $\bar{Y} = \{\bar{y}_1,..,\bar{y}_n\}\in [0,1]$, I want to perform ...
1
vote
0answers
466 views

Feature selection methods comparison

I recently run a project that involves a feature selection step before further pattern recognition. The number of features for our data set is very large and instead of running greedy ...
1
vote
2answers
675 views

The curse of dimensionality? (linear SVMs)

How do you know whether you suffer from it? Let's suppose I have a 2 class problem - 2000 training examples and 30 features. While it works good for the most part, sometimes I get edge cases that ...
1
vote
1answer
72 views

Algorithms/methods to create more features of a limited amount of features?

So, let's suppose that I have a set of 20 features - some of them are continous and some of them are binary. Is there an algorithm/method/solution to create more features ( combine/transform ) those ...
2
votes
1answer
757 views

How to interpret this cross-validated sparse LDA figure using CARET package?

Training data with $p$ =11 predictors and $n$ =165 with 4-class problem was cross-validated (5 times repeated 10-fold CV) using the sparse LDA (aka SDA) using caret ...
1
vote
1answer
83 views

SVM - combining binary and continous representation of the same feature?

How would this influence the accuracy of the SVM model? Let's suppose that I have one variable which max value is 100 and minimum is 0. Currently, I send it to SVM as a single continuous feature, ...
1
vote
1answer
39 views

Significance Test for Comparison of Variables

I am not sure how to ask this question without giving an example. I am trying to measure the "cleanliness" of office buildings. I have two variables that try to measure this. Variable one is a ...
2
votes
1answer
506 views

Using principal component analysis (PCA) for feature selection in regression [duplicate]

I have a dataset $D$ made of $m$ samples and $n$ features with $n \gg m$. For each sample I have a score $s$ which I would like to be able to predict. As the number of features is very high (compared ...
2
votes
2answers
4k views

Mixing continuous and binary data with linear SVM?

So I've been playing around with SVMs and I wonder if this is a good thing to do: I have a set of continuous features (0 to 1) and a set of categorical features that I converted to dummy variables. ...
2
votes
2answers
317 views

Selecting a feature modeling approach for text classification

I am new to text processing. Currently I am trying to determine which type of feature vector I need for a classification problem. I am mainly deciding between binary feature modeling and ...
2
votes
1answer
112 views

Are there any techniques that quantify the importance/signification of individual attribute values of a particular data point?

Are there any techniques that quantize the importance of individual attribute values in a particular data point, in terms of the attribute's overall importance/signification/contribution to the ...
5
votes
1answer
685 views

Explain steps of LLE (local linear embedding) algorithm?

I understand the basic principle behind the algorithm for LLE consists of three steps. Finding the neighborhood of each data point by some metric such as k-nn. Find weights for each neighbor which ...
0
votes
2answers
96 views

Choosing the best featureset for prediction

I have this N sets of features F each with $F_i$ number of features. All the feature sets have 20000 examples and we have 20,000 labels for them. Lets say feature set $F_1$ has 10 features and ...
2
votes
2answers
471 views

How to calculate number of features based on image resolution?

Just covered Andrew Ng's Non-linear Hypothesis of Neural Netowrks, and we had a multiple choice question for determining number of features for an image of resolution 100x100 of grescale intensities. ...
20
votes
4answers
1k views

How can SVM 'find' an infinite feature space where linear separation is always possible?

What is the intuition behind the fact that an SVM with a Gaussian Kernel has infinite dimensional feature space?
2
votes
3answers
304 views

improve precision in text classification

I am working on binary text classification using sklearn: The length of each sample is not high (~ 200-500 characters) I use TF-IDF to get important words as TfidfVectorizer(sublinear_tf=False, ...
2
votes
1answer
217 views

Purposeful selection and confounding

I conducted purposeful selection as outlined in Jewell's Statistics for Epidemiology. The log likelihood tests showed covariates, which I considered to be confounding though not significant in the ...
1
vote
0answers
61 views

Inferring dimension weight in a mapping from a triangle to a distribution over its vertices

I have a dataset $(y_i, \mathbf{X}_i)$, where $\mathbf{X}_i$ is a $3 \times n$ matrix of reals and $y_i$ takes a value in $\{1, 2, 3\}$. Essentially, $y_i$ represents a "selection" of the row vector ...
0
votes
3answers
109 views

Classification performance and the feature set selection

I am now working on a classification problem. The generated feature set can be separated into two group. I did a comparison study: use all of the features; use the features of group 1 only; and use ...
2
votes
1answer
149 views

penalized regression applications in epidemiology

I am seeking advice on penalized regression models for selecting covariates in epidemiological studies. A difficult tasks I face is feature selection while still attempting to account for confounding ...
2
votes
1answer
59 views

Should I exclude predictor variables if used to create a new one?

I have a dataset that includes race, gender, income, and family size. In addition, a variable for "sliding fee scale" tier is included, which is determined by income and family size. Should income and ...
0
votes
1answer
189 views

Steps followed when Binary logistic regression when both dependent and independent variables are binary

I had set of binary variables. To apply logistic regression, I have checked association between dependent and independent variables and considered only those independent variables in the model which ...
7
votes
4answers
13k views

Term frequency/inverse document frequency (TF/IDF): weighting

I've got a dataset which represents 1000 documents and all the words that appear in it. So the rows represent the documents and the columns represent the words. So for example, the value in cell ...
23
votes
4answers
2k views

When wouldn't I use LASSO for model selection?

Assume that you need to build a linear model to make predictions for new observations, and that there is uncertainty about which subset of variables should be included in the model. You are only ...
1
vote
2answers
441 views

CV on training set with feature selection

I've got a problem with CV on feature selection. I've used a method, but I don't know it's correct... I split my data into 70% training set and 30% test set I work now with my training set. I do on ...
3
votes
1answer
1k views

Step-wise feature selection with caret

can anyone direct me to a package/commands in R for performing step-wise feature selection, preferably using the caret package. I have already used linear ...
0
votes
1answer
237 views

Feature selection and cross validation

I'm working on a project and I would like to know if the following strategy is good/correct. Sorry if this is a basic/stupid idea (I'm new to this). The input is a dataset with 2.500 features and ...
1
vote
0answers
115 views

How to explain difference of importance between feature selection and model quality?

I have a data collection with a mixed feature set consisting of both numerical features and text features. The number of numerical features is quite small, i.e., 6, comparing to the number of text ...
0
votes
1answer
193 views

Random forest importance differs between rf$importance and importance()

My model is working ok (the AUC is 0.7) but the importances from a randomForest run for my binary classification problem differ depending on how I retrieve them. Is ...
0
votes
1answer
43 views

Predictive features with high presence in one class

I am doing a logistic regression to predict the outcome of a binary variable, say whether a journal paper gets accepted or not. The independent variable or predictors are all the phrases used in these ...
1
vote
1answer
103 views

Numeric example of data for special case of stepwise linear regression

Stepwise Regression works as follows if I'm correct: fit the initial model add the variable which has its f-stat larger than a in-threshold and repeat step 2. if there are no candidates to enter - ...
2
votes
3answers
933 views

How to divide feature set for selection and training

I have training data with 260 observations that have a total of 7 classes. Each observation has 120 features. I applied feature selection based on the Bhattacharyya Algorithm and got the top 40 ...
1
vote
1answer
63 views

finding an optimal subgroup of binary indicators

My dependent variable is continuous variable that measures the (potential) success of a person in some activity. I have hundreds of binary indicators, each indicates about the existence of a specific ...
3
votes
0answers
136 views

Does full subset selection suffer from the same handicaps as stepwise regression?

Let's assume $p$ potential predictor variables $X_1,...,X_p$ and a single dependent variable $Y$. Now I evaluate the performance of all possible linear models considering all possible combinations of ...
1
vote
0answers
308 views

Cross validation for variable selection and coefficient shrinkage?

Is cross validation an appropriate technique for variable selection and regression coefficient shrinkage? A former colleague of mine used 10-fold CV to compare the regression coefficients from the ...