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

learn more… | top users | synonyms (2)

2
votes
1answer
74 views

SVM Classifier: proper process

I'm working on a classification system of mine, but am needing help with the proper process order. Specifically, I'm using LibSVM and a range of feature sets extracted from my data. I'm wondering, ...
0
votes
1answer
103 views

determining how “important” a feature is in predicting a target in decision trees

Random forests allow us to compute a heuristic for determining how "important" a feature is in predicting a target. This heuristic measures the change in prediction accuracy if we take a given ...
1
vote
1answer
49 views

Do Neural Networks need “compound” features?

Apologies if I haven't got the terminology quite right. I have a question about Neural Networks, and I'm not sure exactly the best way to ask it! Hypothetically, let's say I have a dataset of houses ...
0
votes
1answer
175 views

What does the varImp function in the caret package actually compute for a glmnet (elastic net) object

I am fitting an elastic net model with glmnet via the caret package with 189 predictors and a binomial criteria (a,b) ...
0
votes
0answers
15 views

How to predict the predict upper bound of a learning algorithm?

I am selecting features for a Logistic regression classifier. I have tested a lot of feature selection algorithms. however, it seems that there exist a fixed upper bound AUC value for a fix feature ...
0
votes
0answers
24 views

What is the best practices as to cropping the positive examples for training hog classifier?

I've been reading up on the literature on HOG training, but I can't figure out what the best practices are for cropping the positive examples. Question: Do you crop tight or with margins on the ...
2
votes
1answer
109 views

K-means clustering feature selection

I have a set of English and foreign language documents that I would to perform k-means clustering on to find document groups by topic. These documents are concatenated social media comments for ...
0
votes
0answers
24 views

Bayesian network overfit - number of features and examples

For a dataset consisting of 150 examples (mostly binary features) what would be the number of features needed so that a Bayesian network doesn't overfit? I know there is no exact answer and I've ...
1
vote
0answers
39 views

Poisson model with fractions

I have a simple website with a home page that has 5 different images on it. All images have a fixed set of 'features' associated with them (size, color, position etc.,). When a visitor comes to the ...
0
votes
1answer
106 views

Selecting variables in multiple linear regression in R

Consider that we have a problem with 4 variables (y, x1, x2 and x3) and we want to do a multiple linear regression model. As we need to know which variables are the most important in the problem, we ...
1
vote
1answer
14 views

Multi Categorical Features vs multiple Features for categories

Say I am discretizing continuous data based on percentiles. (I realize this is generally frowned upon, but I am doing this for the sake of experiment) I am trying different percentiles, eg breaking ...
0
votes
1answer
100 views

combining multiple classifiers common features

Can multiple binary-classifiers be combined to produce a final output if their feature sets have some common elements? How will this influence the accuracy?
0
votes
0answers
30 views

Collection of features that correlated with strength and elongation of nonwoven fibers

I'm working on a regression problem which estimates strength and elongation of a non woven fibers by using some features collected from processed images of the fibers. For this study I've collected 50 ...
0
votes
1answer
31 views

Using a set as a feature in decision tree classification

I'm faced with a data set where one of the features is a set of 4-5 categories (this number of categories isn't constant). I need to use this feature for building a decision tree. I searched online ...
3
votes
1answer
62 views

Comparing classification results from models selected differently with Kruskal-Wallis

For a seminar at university I'm reviewing a paper that proposes a new feature selection algorithm. The authors also evaluated their algorithm by applying it beside two other feature selection ...
1
vote
1answer
34 views

Getting less number of features in weight vectors as were provided for SVM

I have trained a SVM with 18881 features and wanted to know the ranking of features. I tried the method given at http://stackoverflow.com/questions/7390173/svm-equations-from-e1071-r-package for it ...
0
votes
0answers
158 views

K Fold Cross Validation, Variable Selection in LDA

I'm currently working on a multi-class classification problem and I attempt to use lda for the same. I have 2 questions here. 1) Is it possible to perform k-fold ...
0
votes
1answer
152 views

Large data variable selection

I'm looking for some methods of variable selection on large datasets.The number of variables are around 30-40, but the number of observations is quite large (around 36000000) Any methods which I ...
0
votes
0answers
33 views

Contribution to the components of a Gaussian mixture by data features

My question is about modelling data with a GMM using EM. One can split the mean and variance of each component into parts as well when working with data with multiple features. My question is what ...
0
votes
0answers
46 views

How to check that the selected features are not overfitting

I'm trying to select best features for sentiment classification for a set of reviews, and using penalized SVM and Logistic regression to perform such task. by basically iterating over different ...
0
votes
0answers
14 views

Performing interaction with a very significant predictor drives down p val of other predictors. But does it makes sense?

I'm observing a phenomenon that I can't understand. I have a linear regression setting with categorical vars. A couple of these elicit an highly significant coefficient and low p values. When used ...
1
vote
1answer
20 views

How to find the cause of defect in a process

Suppose a product A undergoes a certain process. This product A is produced at a rate of 8000 per month and out of those in 75 cases defects are generated. In the data set, I have rows corresponding ...
2
votes
1answer
71 views

Difference of variable selection and importance estimation

Isn't variable importance estimation a necessary prerequisite for variable selection? Is there any use case where you want to select non-important variables for your model? So, why is variable ...
0
votes
0answers
67 views

Multple linear regression, adding one predictor with almost perfect fit make others irrelevant

I found something interesting while playing with some data and linear regression. I built a regression with various predictors, more or less correlated with the outcome. Then I added one predictor ...
0
votes
1answer
34 views

Feature selection for unknown parametric model

Suppose one has about 500 points of 50 dimensional data that one knows a priori is derived from a parametric model (perhaps with some outliers). Does using this knowledge help in feature selection? I ...
2
votes
2answers
97 views

Do I need to take out any predictors from multiple regression if I put in some principal components as additional predictors?

I have an assignment which involves one area-level dataset made of $366$ scale variables. I have to perform PCA, compare it with rates of an additional response variable $X$, and comment on its face ...
0
votes
1answer
137 views

How to interpret random forest importance numbers

I ran randomForest in R package using 7 predictors variables (x1 to x7). I repeated the test with 4 dependent variables (y1 to y4). The importance numbers (IncNodePurity) are plotted in following ...
1
vote
2answers
181 views

How to build a predictive model with a billion of sparse features?

I am making a model to learn a dataset which has a big feature number and sparse samples (I am planning to use logistic regression). The feature number can be as big as 1,000,000,000. It is sparse ...
3
votes
0answers
33 views

Development data set for feature engineering and data exploration

I dont hear this being talked about much: If you want to engineer features and visually explore the data, should you do this on a development set separate from the training and test set? If ...
0
votes
0answers
30 views

Machine Learning: Potential Reasons of Precision Change after New Features are Added

My baseline model uses 10 features $[f_1, f_2, \dotsb, f_{10}]$. Now I have two new features $f_{11}$ and $f_{12}$. New models that use either $[f_1, f_2, \dotsb, f_{10}, f_{11}]$ or $[f_1, f_2, ...
0
votes
0answers
30 views

selection of features with Weka

I have a question and I hope that you can help me: I have a bilingual text(source language and target language). I extract from this text the best source phrase and the target phrase related to this ...
0
votes
1answer
62 views

Chi-squared Vs Mutual information

Is chi-squared feature selection better than Mutual information based feature selection mechanism?
0
votes
1answer
41 views

Effect of combining features on classification

I have 2 string features F1 and F2 based on which I am trying to perform classification. I have two choices, either to use the ...
0
votes
1answer
50 views

After Clustering, how can I evaluate which features had the biggest impact?

I've just performed unsupervised clustering (using DBSCAN) on a dataset for which I have no expert knowledge on. I'm interested in working out which features had the greatest impact on my clustering. ...
2
votes
2answers
93 views

Backward feature selection with CV model selection

I am thinking about doing the following to a data set with $N$ samples and $m$ features 1) Train using semi-supervised learning and cross validate on labeled data using LOO-CV to select the best ...
0
votes
0answers
69 views

Advice for feature selection or feature extraction with semi-supervised learning

I am trying to solve a semi-supervised learning problem using LaplacianSVM. However, before applying LapSVM I would like either to perform feature selection or feature extraction. Furthermore, after ...
0
votes
0answers
10 views

Modeling 2-D data as vector

As part of my class project i had to use decision tree classification on a training set which contains a set of matrices where each row is a vector recorded at a particular time stamp and each row ...
0
votes
0answers
15 views

Can unsupervised feature learning be used to develop features reflecting patterns in human relationships?

Unsupervised feature learning has been used to learn features for objects and action classification and for emotion detection in speech. My questions are: (1) Is there any existing research ...
1
vote
1answer
66 views

Variable selection and validation dataset

According to Hastie & Tibshirani, we shouldn't use validation datasets to do variable selection; otherwise, we will overestimate the model fit. However, it seems quite often to select variables ...
0
votes
0answers
72 views

regarding using Lasso and Random forest based on the variable selection result coming from other processes

After the process of data exploration process and discussion with client, we set up a set of variables as follows: T1, T2, T3, T6, T8, T2*T3, T1*t6 During ...
0
votes
1answer
78 views

NLP tokenization for building feature vector

I am trying to match new product description with the existing ones. Product description looks like this: Panasonic DMC-FX07EB digital camera silver. These are steps to be performed: Tokenize ...
1
vote
1answer
182 views

How to handle missing data in a small $n$ large $k$ machine learning scenario?

I have a sample size $N=130$ and $1000$ variables. I am using machine learning techniques (SVM) for analysing the data. Some variables in the dataset have values that are so huge that they must be ...
1
vote
1answer
138 views

Regarding the different variable selection result between regression modeling and random forest

I build a prediction modeling using both regression and random forest. ...
1
vote
0answers
18 views

Classification with two different dataset

I am working on a cancer classification model.Task is ,I am initially given a data set of 500 people and 1000 features.These people are given some kind of treatment(say Treatment 1). Some people are ...
4
votes
1answer
78 views

Placement of earlier features in more complex features in CNN

I'm trying to understand convolutional neural networks better. I've been doing different tutorials, but there are some basics concerning how the hidden units represents features that I really would ...
0
votes
0answers
12 views

selection variable algorithm based on conditonal mutual information

Is there anyone who understand the selection variable algorithm based on conditonal mutual information proposed by Fleuret 2004? http://sci2s.ugr.es/keel/pdf/specific/articulo/Fleuret_Fast_2004.pdf ...
1
vote
2answers
76 views

How to best to use Continuous value features with discreet values for logistic regression based binary classification problem

This is related to Minimisation algorithm for a mix of discreet and continuous parameters? I am trying out logistic regression to solve a binary classification problem. Though I am feature-scaling ...
0
votes
0answers
14 views

Non-linear auto-regressive model - preselection of relevant columns

Let us consider a dynamic system with nonlinear auto-regressive evolution such as $$ x_{t} = f(x_{t-1},x_{t-2},\dots,x_{t-d})+\epsilon_t $$ where $x_t\in\mathbb{R}^n$ is vector and $\epsilon_t$ is a ...
5
votes
1answer
100 views

“…if the data is linearly separable”

I keep hearing this phrase as a precursor to many algorithms, but I am not sure how exactly one goes about finding out if the data is indeed, linearly separable. Of course, if the data has ...
1
vote
1answer
180 views

Grid search for SVM parameters; is this is really how it is done?

Suppose I use nested 10-fold cross-validation with SVM. So, the inner-most loop will go around 100 times. Now, suppose I use a gaussian radial basis kernel function, which needs the parameter sigma. ...