1
vote
1answer
18 views

Quantifying applicability domain for predictive models?

I have a big dataset and I want to build a classification model (svm, rf, ann etc.). Then I split the original dataset into training set and test set. I build the model using training set. After it ...
1
vote
1answer
29 views

Computational Complexity of Prediction using SVM and NN?

I've seen answers discussing the complexity of training SVMs and neural nets, but how about for predicting new responses once a model has been trained? For context, I'm working on an app that should ...
1
vote
1answer
25 views

Hidden Markov Model: Predict observation sequence from state sequence

Given a transition matrix, starting probability, means and covariances Is it possible to predict the most likely observed sequence for a given state sequence using the above details? If yes, how? ...
0
votes
0answers
23 views

Literature for prediction models where each training example has a different amount of data?

This could be a machine learning question as much as a statistics question, but I think this is the best place to put the question. Here are three different examples of problems where each ...
0
votes
0answers
46 views

Time Period Predictive Modeling

I have been implementing some classification algorithms (Naive Bayes, SVM etc) recently on the iris data sets to get head start into the data science field. I enjoy working on machine learning ...
0
votes
0answers
19 views

Proper way to determine attribute feature selection's smaller subset based on result metrics

Overview My goal is to predict survival of an instance for five different time periods (binary attribute). I have a 100,000-instance dataset with 40 attributes and I want to reduce the attributes ...
1
vote
1answer
308 views

Interpretation of a WEKA result buffer - confusion matrix and performance

I want to know how to get several performance measurements of a generated WEKA model. Note that I am predicting a two-class variable, Alive or ...
8
votes
2answers
128 views

Coupling time series information from sources with multiple spatial resolutions/scales

I have many satellite raster images available from different sensors. From these, the coarser ones have a very abundant temporal resolution. The medium resolution rasters tend to have less acquisition ...
0
votes
0answers
30 views

Machine Learning techniques for classifying response from several continuous factors

I have some data with rows corresponding to male-female pairs, with 8-10 continuous factors (that have been unfortunately anonymized) that apply to both the male and female, and whether they became ...
1
vote
1answer
90 views

Learning curve shows decreasing accuracy

I'm working on a random forest classifier with 10-folds CV to aestimate the hyperparameter 'mtry' (chosen by maximizing AUROC). I decided to pre-split the training set in 8 samples equals in size ...
0
votes
0answers
25 views

Assessing whether 2 models differ “substantially” (though equal RMSE), to improve prediction

In preparation of stacking several models, I would like to gather some heterogeneous models. I am not sure, but expect that the stacking will improve when I have very different models, though all have ...
0
votes
2answers
83 views

A multi-label classification for tagging short text

I am fairly new in the area of text mining and want to practice my skills a little. I have the following task at hand which I want to work on. I have a large list of short texts (~100.000) and every ...
4
votes
1answer
126 views

Why discriminative models are preferred to generative models for sequence labeling tasks?

I understand that discriminative models, such as CRF(Conditional Random Fields), model conditional probabilities $P(y|x)$, while generative models, such as HMM(Hidden Markov Model), model joint ...
0
votes
2answers
50 views

Predictive algorithm validation

In putting a binary 1/0 predictive algorithm into production, what are the consequences where only the positive (1) predictions are checked, meaning only true or false positives are detected, and then ...
2
votes
1answer
293 views

Hidden Markov model for event prediction

Question: Is the set-up below a sensible implementation of a Hidden Markov model? I have a data set of 108,000 observations (taken over the course of 100 days) and ...
0
votes
0answers
33 views

Looking for a list of modelling techniques for continuous response with missing values in predictors

I am looking for a list of modeling algorithms (as a package in R) that can accept: Continuous or categorical predictors Continuous response Can effectively ...
7
votes
3answers
680 views

Cross-validation or bootstrapping to evaluate classification performance?

What is the most appropriate sampling method to evaluate the performance of a classifier on a particular data set and compare it with other classifiers? Cross-validation seems to be standard practice, ...
2
votes
0answers
34 views

Using sequential observations to perform online prediction

I'm trying to perform predictions from a sequence of events. My problem is this: Data collection: Suppose you can continuously observe a person sitting in a library. You take note of every time that ...
0
votes
0answers
44 views

Performing online prediction from sequential observations

I am trying to perform some predictions from a sequence of observable events. My problem can be abstracted like this: Data collection: Suppose you can continuously observe a person sitting in a ...
6
votes
1answer
488 views

How to predict new data with spline/smooth regression

Can anyone help give a conceptual explanation to how predictions are made for new data when using smooths /splines for a predictive model? For example, given a model created using ...
2
votes
0answers
40 views

Prediction errors for Adaboost

I understand how the adaboost algorithm works to produce a prediction of a class, however one thing I haven't seen is how to get a measure of accuracy for that prediction. For example, if I fit a ...
2
votes
1answer
69 views

What is the difference between the values in the 'fit' attribute of a gbm object and values computed by gbm.predict?

My intuition is that the fitted values and predicted values of a gbm object should be identical. But in this example with just one tree, the values are different: ...
2
votes
0answers
88 views

Measuring parameter sensitivity and variability (standard-error) in k-fold cross-validation

I mainly use k-fold cross-validation for parameter tuning and model selection for prediction problems. Now, is there a standard or if not a less-known way to measure the sensitivity of the parameters ...
0
votes
0answers
84 views

High Dimensional Data, Multicollinearity, and Skewed Dependent Variable - a good approach?

I have a regression problem that has three key issues I am trying to tackle. The first is the data is a p >> n problem, e.g. I have about 1500 features but only about 150 examples. The second is that ...
2
votes
2answers
186 views

What machine learning techniques can, once trained, generate prediction despite some missing inputs?

I have a training set where the inputs & outputs are all present, but I suspect that in the data where I want to do prediction, I will occasionally encounter scenarios where a small fraction of ...
2
votes
1answer
156 views

When to Log/Exp your Variables when performing Linear Regression?

I'm doing regression using Random Forests for predicting prices based on several attributes. Code is written in Python using Scikit-learn. How do you decide whether you should transform your ...
6
votes
2answers
247 views

Why do categorical predictor variables in regression need to be recoded as multiple predictors?

I'm learning about machine learning using Python's library scikit learn, and in their tutorial here they mentioned about a categorical variable color which can have ...
3
votes
2answers
266 views

Why is Hedonic Regression used instead of Linear Regression

Why is Hedonic Regression used (especially in housing prices) instead of Linear Regression? There do not seem to be any libraries in Python (and R) for Hedonic regression, is it too niched a ...
1
vote
4answers
181 views

Measuring representativeness of a sample using covariates

I was provided with quite a small sample of labeled (variable of interest) observations to train a model to predict unlabeled observations. All the observations are associated with many covariates. ...
2
votes
1answer
109 views

What exactly is the equation for SVM classification for new example?

I understand that in the case of Logistic Regression, we simply multiply our weights with Input example for classification. But what exactly is the equation that we calculate in the case of SVM to ...
8
votes
2answers
208 views

Is there overfitting in this modellng approach

I recently was told that the process I followed (component of a MS Thesis) could be seen as over-fitting. I am looking to get a better understanding of this and see if others agree. The objective of ...
4
votes
2answers
136 views

Predicting chemical property (Boiling Point) from a SMILES string

I was trying to develop a model for predicting Boiling Points (BP) given a chemical name. One good and unique (ok, almost) way to encode a name is the SMILES notation string. The details of the ...
1
vote
0answers
50 views

Prediction with intervals as the independent variable

I have sample data that maps intervals to a number: [3,7] => 1 [6,8] => 2 [6,13] => 3 [7,10] => 3 [10,13] => 4 The dependent variable's values ...
6
votes
1answer
493 views

Predicting football match winner based only on the outcome of previous matches between the two teams

I'm a huge football (soccer) fan and interested in machine learning too. As a project for my ML course I'm trying to build a model that would predict the chance of winning for the home team, given the ...
3
votes
0answers
151 views

Shifted intercepts in logistic regression

I have a question about the effects of shifting the intercept in a logistic fit on the mean of a particular transformation of the scores. Here is the notation I will be using for the question. The ...
8
votes
5answers
352 views

Does preclustering help to build a better predictive model?

For the task of churn modelling I was considering: Compute k clusters for the data Build k models for each cluster individually. The rationale for that is,that there is nothing to prove, that the ...
4
votes
0answers
421 views

High-dimensional Regression Datasets [closed]

Am looking for pointers to publicly(online) available high-dimensional regression datasets for evaluating my research work. By high-dimensional, am looking for regression datsets with the number of ...
1
vote
1answer
121 views

Robust regularized regression

I've been using elastic net implemented in R (via glmnet) for some modeling, but I was wondering, due to the number of outliers in my data, if there was some sort of modeling approach for regularized ...
2
votes
1answer
74 views

Effect of varying outcome duration in longitudinal studies

I have a supervised classifier model (regularized discriminant) which predicts the probability of an event occurring within two years. This model was developed using sensor data measured from a ...
1
vote
0answers
60 views

Distance correlation and prediction

If the distance correlation (ref. Gabor J. Szekely) $R_n(X,Y)>R_n(Z,Y)$ would the expected generalization error of a prediction model over $(Z,Y)$ be lower than $(X,Y)$ in predicting $Y$, where ...
1
vote
0answers
227 views

Gaussian process - dimensionality reduction

I saw a method for dimensionality reduction for the squared exponential covariance function (not ARD) whereby one uses a $G\times D$ projection matrix $P$ ($G < D$, $D$ = dimension of the inputs) ...
0
votes
1answer
829 views

Gaussian Process regression for high dimensional data sets

Just wanted to see if anyone has any experience applying Gaussian process regression (GPR) to high dimensional data sets. I'm looking into some of the various sparse GPR methods (e.g. sparse ...
1
vote
0answers
179 views

Guassian Process Regression - feature selection

I'm using guassian process regression to do some modeling. One issue I'm encountering is feature selection for some of my models, which often have many relevant features. I'm not sure what the best ...
3
votes
2answers
963 views

Why does Lasso do better than SVM?

This is a soft-question: I have been evaluation various regression techniques over a regression dataset that I have. I am surprised by the fact that cross-validated RMSE of Lasso is better than SVM ...
2
votes
2answers
202 views

SVM and non-linear predictive models - feature selection

Just throwing out a general question. What do people think of applying feature selection methods when using SVMs to build predictive models? I understand that SVM have built in regularization with how ...
2
votes
0answers
138 views

Robust Support Vector Regression - robust to outliers

I've been reading/looking around for literature on support vector regressions that are relatively robust to outliers. I understand that standard SVRs can be significantly influenced by a few large ...
5
votes
3answers
642 views

Support vector regression on skewed/high kurtosis data

I'm using support vector regression to model some fairly skewed data (with high kurtosis). I've tried modeling the data directly but I'm getting erroneous predictions I think mainly due to the ...
6
votes
2answers
196 views

Incorporating a treatment into a classification scheme

I have about 400 pieces of silver of different geometric dimensions. They were assigned to six groups and each group went through a series of stress tests, such as bending, pulling, putting in fire ...
1
vote
1answer
214 views

SVM parameter selection with NM simplex (or other algorithms)

I'm having some trouble getting the NM Simplex to find a good minimum for selecting hyperparameters of a rbf SVC. Not only am I tuning the 2 SVC parameters (C and gamma) I also have five class weights ...
2
votes
3answers
131 views

Multi task learning

I have a dataset where all observations are measured several times and reported outcomes correspond to those measurements. In other words, my set of data points looks like $\{x_i, y_{i_1}, y_{i_2}, ...