The tag has no wiki summary.

learn more… | top users | synonyms

0
votes
0answers
7 views

Confused about cross validation for model stacking

I'm reading section 8.8 of Elements of Statistical Learning, and though I keep reading the section on calculating the ensemble weights I'm missing something. It says that the stacking weights are ...
0
votes
0answers
13 views

RandomGLM: data (responses) is not read as numeric [on hold]

I'm relatively new to R and I have a problem with randomGLM for prediction of a continuous variable. I read in four datasets which are comprised of a training ...
0
votes
0answers
10 views

What techniques are used to prevent overfitting in DSN

A Deep Stacked Network (DSN), is a ensemble learner, which roughly works by training a single hidden layer neural network on the inputs and target outputs, then training another which takes an input ...
0
votes
0answers
35 views

credit scoring - fraud scoring

I have been asked to build a credit scoring model and we are relying on several Machine Learning API, in order to build our feature vectors. One of these API is MinFraud. However, as they provide us ...
1
vote
0answers
21 views

What's the potential reason that by combining two feature sets the performance of random forest dropped?

I am building random forests on high dimensional, sparse, and class unbalanced training datasets (around 500 - 5000 examples) using two different feature sets. I did stratified 10-fold cross ...
1
vote
0answers
18 views

Assessing predictor contribution to model output

Many of machine learning methods are considered as "black boxes". Examples of such methods are SVM, Neural Networks, Random forests etc. One may apply sensitivity analysis techniques (as described for ...
1
vote
2answers
78 views

Analysis for checking if an Ensemble model is a better fit for a dataset than Primitive model

I have a dataset and have the option to apply either GLM (primitive) or a Random Forest (ensemble). So far the Random Forest is giving way better results than the GLM. As it is generally believed that ...
0
votes
0answers
12 views

In ensemble predictions, what is reliability diagram?

I have recently heard the term reliability diagram used with regards to the analysis of ensemble predictions. What does it show and how is it calculated?
0
votes
1answer
31 views

What statistical analyses should one perform on ensemble forecasts (given a measurement)?

I have an ensemble of time-series predicting a scalar variable. Additionally, I have a measurement time series of this scalar variable. Which statistical analyses could and/or should I perform to ...
0
votes
1answer
133 views

Suggestions needed about classifier fusion

I'm working on a classification problem which involves two classifier to observe a single event. I'm providing a high level description of the problem without going into the technical details (the ...
4
votes
0answers
25 views

Polling vs averaging in Random Forest models

Why is it that for Random Forest we take the average vote from each classifier in the ensemble rather than the average probability from each classifier in the ensemble? Is there theory behind why ...
1
vote
0answers
15 views

Proper way to stack models when some models aren't always applicable?

Suppose you have two (or more) models that you want to ensemble together. However, some of the models are trained specifically on very specialized subsets of your data. If you do the stacking with a ...
0
votes
0answers
57 views

Stacked Generalization Ensemble Algorithm for regression

I am using stacked generalization(Rupert 1992) for combining multiple(8) heterogeneous base learners for regression. What I understand from the pseudo codes that Train the 8 learners on 8 instances ...
1
vote
0answers
28 views

Ensemble model performs better with worse performing consitutent models?

I have a forecast model I am developing that uses some very unreliable input data, missing data (due to sensors or comms failures) is the rule, not an exception. The quantity being forecast is a daily ...
2
votes
1answer
200 views

Why not always use ensemble learning?

It seems to me that ensemble learning WILL always give better predictive performance than with just a single learning hypothesis. So, why don't we use them all the time? My guess is because of ...
0
votes
1answer
46 views

What is this ensemble learning technique called?

For example, I trained two models: one with SVM and one with KNN. Final Prediction = 0.4*KNN + 0.6*SVM Is this considered blending?
0
votes
1answer
216 views

What is the equation for random forest?

I need an equation for random forest so that I can score fresh data I receive every week, based on beta estimates I got after building model using this ensemble methodology. Every week I do not want ...
0
votes
1answer
27 views

What method to use for cluster identification ?

This question is from a confused novice. I have a data set with where each point is located in a 2-D space defined by two objectives (say, X and Y). I wish to identify a set of points from this space ...
4
votes
2answers
198 views

Ensembles of Ensembles?

I like the idea of ensemble learners, especially Bagging, but I always wondered as why they are not the most powerful learners since they have a clean motivation. I don't have the answer to that ...
0
votes
0answers
17 views

combining classifiers trained on different data

Could you please recommend methods for combining classifiers, which were trained on different patients? For example, there are 10 patients and for each of them I trained a binary classifier. Now I ...
2
votes
1answer
170 views

k-fold Cross validation of ensemble learning

I am confused about how to partition the data for k-fold cross validation of ensemble learning. Assuming I have an ensemble learning framework for classification. My first layer contains the ...
1
vote
2answers
131 views

Combine decision trees from GBM to reduce output

I am curious if any research has been conducted to efficiently combine trees resulting from a gradient boosting process. I routinely run a process that generates 20 or 30 thousand trees in R. I then ...
1
vote
0answers
52 views

How to combine predictions in ensemble

I am trying to learn more about how to build ensembles of predictions in R and coming to a roadblock, and am hoping one can offer guidance. I often read about people automatically identifying how ...
1
vote
0answers
75 views

What are the latest methods to generate ensembles?

I am working with ensembles, and I'm willing to go deep inside the work. I have historical records of: Observations of one variable Historical forecasts for the same variable For future ...
3
votes
2answers
77 views

Combining models for prediction based on residual performance

I have never read or seen someone do this before, so I wanted to pose the question here. Suppose I fit a basic linear model, $\text{price of house} = \beta_0 + \beta_1*\text{taxes} + ...
0
votes
2answers
77 views

Ensemble of models with different feature spaces

BACKGROUND I have data in which the dependent variable is binary with a highly-skewed distribution: <1% records are 1 (doers), >99% records are 0 (non-doers). I'm using logistic regression to ...
0
votes
0answers
31 views

Building Ensemble Models Practical Question

I understand the general idea of stacking several models and combining them to create a model that might perform better than any of the individual models that it is composed of. My question is ...
2
votes
0answers
59 views

Dissimilarity with earlier features part of cost function

I am using a RandomForest on features (pixels) of images, and I am considering adding cost for "similarity to already other included features" to the cost function. Imagine you have a current RF ...
0
votes
0answers
31 views

Early split decision criteria for fast random (regression) forest estimation

Suppose I am on a node in a $regression$ tree and I am using running estimates of $\sum_{i \in Region_1} (y_i - mean(y_i)_{Region1})^2$ (and the same for Region 2) to determine whether to split the ...
11
votes
3answers
1k views

Can Random Forest Methodology be Applied to Linear Regressions?

Random Forests work by creating an ensemble of decision trees where each tree is created using a bootstrap sample of the original training data (sample of both input variables and observations). Can ...
3
votes
2answers
171 views

Why are Random Forests splitted based on m random features?

I was watching the following tutorial on Random forests and it says that, "at each node, choose random subset of m features and only consider splitting on those features" I do not understand why we ...
0
votes
0answers
34 views

Comparing estimates from different models

How would I compare (posterior) probability estimates of a document from n classifiers trained separately on different datasets (the # of classes in each classifier is different). Is there a way to ...
1
vote
0answers
84 views

Proper splitting of data set for Ensemble methods

I have 10,000 documents. Each document has a label ($Y$) that is either $0$ or $1$ (the 0-1 split is pretty much 50/50 over my 10,000 documents). Each document has 10 fields. Each field can have any ...
5
votes
1answer
133 views

Do ensemble techniques increase VC-dimension?

Techniques like Adaboost use a ensemble of weak classifiers to obtain a "better" classifier. Does(Can) the final classifier have a greater VC-dimension than the weak classifier? An intuitive ...
1
vote
1answer
101 views

Gradient boosting algorithm (steps) question

So, far I have read following regarding boosting: Boosting is an ensemble technique. Train learner sequentially, where early learners fit simple models to the data. Analyze data for errors, that ...
1
vote
1answer
331 views

Matrix Factorization Model for recommender systems how to determine number of latent features?

I am trying to design a matrix factorization technique for a simple user-item, rating recommender system. I have 2 questions about this. First in a simple implementation that I saw of matrix ...
0
votes
1answer
80 views

Using a logistic model on the estimates of several other classification models

I'm working on a classification model that will predict whether a sales opportunity will end up 'won' or 'lost', given various attributes of the opportunity. I've been using my training data to build ...
2
votes
1answer
133 views

Is there a well-defined class of ensemble methods?

Ensemble methodology's main aim is to somehow aggregate or summarize estimates from multiple models. In some cases this is aggregating different bootstrap estimates or Monte Carlo estimates, but ...
6
votes
0answers
649 views

Using LASSO on random forest

I would like to create a random forest using the following process: Build a tree on a random samples of the data and features using information gain to determine splits Terminate a leaf node if it ...
4
votes
2answers
376 views

On combining SVMs

Suppose we have a supervised training set $T=\{ (x_1, y_1),\dots, (x_n,y_n)\}$ where $x_i$ is an example and $y_i \in \{-1,+1\}$ is its label. Further suppose that examples are only observable through ...
1
vote
0answers
65 views

One-against-all probability values into a multiple class probability value?

I have a 10-class classification problem. I've approached the problem as a set of one-against-all binary problems. For each class I've built a MLP neural network that provides a probability estimate ...
5
votes
1answer
137 views

Limits to tree-based ensemble methods in small n, large p problems?

Long time grazer, first time poster. I'm hoping to gather people's opinion on the following: Tree-based ensemble methods such as Random Forest, and subsequent derivatives (e.g., conditional forest), ...
6
votes
3answers
573 views

Ensemble time series model

I need to automate time-series forecasting, and I don't know in advance the features of those series (seasonality, trend, noise, etc). My aim is not to get the best possible model for each series, ...
0
votes
1answer
66 views

Defintion for model diversity?

Two models are diverse if they make prediction errors on different instances. I know there are different measures to quantify diversity, however, I'm looking for formal conceptual definition of what ...
3
votes
1answer
810 views

Looking for examples or alternatives to R RuleFit ensemble package

Does anyone know of any good example code illustrations for the rulefit Rule Based Learning Ensembles package? The documentation is incredibly lacking. I was guided to the package by this paper. If ...
2
votes
1answer
392 views

Which are the most effective clustering ensembles?

In supervised learning, there are some ensemble methods that overcome others significantly (adaboost or random forests to mention some). Few years later, also ensembles in unsupervised learning were ...
3
votes
1answer
208 views

How to determine the best number of weak classifiers to use in adaboost without overfitting the data

I was thinking by using validation but not quite sure how to go with it. Please list some papers or ideas on how. This is for multiclass problem (using one vs all approach). I think each ...
1
vote
1answer
516 views

How to convert multiple ranking scores into a probability distribution?

I would like to create a topic distribution for a document. The current model I am trying to implement is: for each sentence in the document, I am getting a topic assignment with a score, e.g. "1st ...
2
votes
1answer
404 views

Concept of iterations in Adaboost

I can't seem to get my head around "iterations" in Adaboost. Are they analogous to weak classifiers that are used for Boosting? I've seen many examples of Adaboost where a programmers use a Single ...
0
votes
1answer
142 views

Accuracy of classifiers with Adaboost

Does Adaboost ensure that resultant accuracy is more than or at least equal to current accuracies? What happens if Classifier A performs badly and the weights are accordingly updated and the next ...