Tagged Questions

From The Discipline of Machine Learning by Tom Mitchell: The field of Machine Learning seeks to answer the question "How can we build computer systems that automatically improve with experience, and what are the fundamental laws that govern all learning processes?" This question covers a broad ...

learn more… | top users | synonyms

0
votes
0answers
14 views

Real utility of small accuracy improvements in sentiment classifiers

I have lately been reading papers regarding Sentiment Analysis, where most researches report that their improvements made them achieve an increase of 1~2%, or even 0.5% in accuracy compared to ...
-1
votes
0answers
45 views

Ridge regression auto-normalization & CV

Is there a package in R that performs regularized regression (ridge, Lasso, LARS, etc.) that also: unit normalizes (i.e. scales and de-means) independent variables automatically AND uses ...
3
votes
0answers
75 views

Regularization L1-norm and L2-norm empirical study

There are many methods to perform regularization -- L0, L1, and L2-norm based regularization for example. According to Friedman Hastie & Tibsharani, the best regularizer depends on the problem: ...
0
votes
0answers
21 views

Multi Class vs 2 class Naive Bayes

I was wondering what are the implications of using a multi-class Naive Bayes versus a 2 class Naive Bayes (for one against everything). Which technique performs better? I've previously came across ...
1
vote
1answer
77 views

Using Python for building machine learning application

I'm currently using R to find the best approach to solving a machine learning problem. Once I've got the approach sorted, I will need to build this into an application which can be used by end users. ...
2
votes
4answers
116 views

Is cosine similarity a classification or a clustering technique?

In document classification, is cosine similarity considered a classification or a clustering technique? But you need training data with the cosine similarity for creation of the centroid right?
0
votes
0answers
172 views
+50

Asymmetrical selective sampling for linear classification

I've got a online classification problem where I predict a class label {+1, -1} for an object and then show it to a user to get a real label. My task is to minimize a number of -1 objects shown to a ...
0
votes
2answers
50 views

Improve SVM performance on data with missing features and outliers in R

I'm trying to learn R for ML purposes, and right now i'm building classifier for my data (10 dimensions, ~400 elements, 2 classes), which have some outliers in it, and a lot of missing values. I'm ...
1
vote
0answers
27 views

One class classification with LIBSVM in Weka

I have a dataset on a particular domain and I want to do a one-class classification with LIBSVM (wrapper) in Weka. I have trained the classifier, but the problem is, when I test it with a different ...
4
votes
2answers
60 views

List of likelihood-based classification techniques

This is a basic statistical pattern recognition question. I'm aware of LDA classification, Naive Bayes Classification techniques which give output as a likelihood (of data belonging to a certain ...
2
votes
0answers
20 views

Evaluating recommender systems where ratings are only 0 and 1

I'm analyzing a set of news articles and user libraries. User library is the set of news articles shared by one user. Obviously, the rating is 1 (the article is in user's library) and 0, otherwise. I ...
3
votes
0answers
29 views

Discerning the best model for a problem

This is a vague question. I will do my best, I think it has definite answers. I am hoping for answers of the form "Read book x, learn this specific topic, read this paper/s". What is bothering me is ...
1
vote
0answers
37 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 ...
0
votes
4answers
100 views

Kernel Selection

I am not an expert in SVM and kernel, so please excuse me if I ask stupid question. Actually, first I want to know how to analyze a dataset to discover its pattern. And second, how can I select ...
3
votes
1answer
38 views

Is there a method for constructing decision trees that takes account of structured/hierarchical/multilevel predictors?

Is there a method for constructing decision trees that takes account of structured/hierarchical/multilevel predictors, that would allow me to impose domain knowledge or constraints on interactions for ...
1
vote
3answers
71 views

Expected best performance possible on a data set

Say I have a simple machine learning problem like a classification. With some benchmarks in vision or audio recognition, I, as a human, are a very good classifier. I therefore have an intuition on how ...
0
votes
1answer
24 views

Cluster analysis with ties issue

When I perform cluster analysis in SAS, the SAS log sometimes return a warning something like this: WARNING: Ties for minimum distance between clusters have been detected at 4 ...
0
votes
1answer
42 views

Help on SVM for road image processing

I am new to SVM. I would like to use SVM to classify road image into two distinct region i.e. drivable region and non-drivable region. I really don't know how to go about this i.e. specifying the ...
2
votes
1answer
56 views

How to detect feature value combinations with outstanding win percentages

I am building ruby on rails (ActiveRecord) horse racing application. One of the features that I would like to include in the application is the ability to identify meaningful "Angles". For example, ...
2
votes
1answer
73 views

Several questions about statistical financial timeseries models from “machine-learning person”

In order to explain why I have those stupid question you'll find below I have to say that I am more a machine-learning person. While I worked on problems in bioinformatics everything was fine. When I ...
14
votes
2answers
146 views

Combining classifiers by flipping a coin

I am studying a machine learning course and the lecture slides contain information what I find contradicting with the recommended book. The problem is the following: there are three classifiers: ...
1
vote
0answers
20 views

Modification of "corrected repeated k-fold cv test” when also averaging Random Forest results across multiple sampling seeds?

I would be very grateful for any ideas concerning the following problem. I would be even more grateful if someone could point me to a literature reference proposing a solution for a scenario like the ...
2
votes
1answer
44 views

Should I use decision trees to predict user preferences?

I'm designing a web service that will predict and recommend new items a user might like based on their expressed preferences on previous items (simple thumbs up/down interface). I was told to look ...
7
votes
2answers
212 views

Choice of K in K-Fold cross validation

I've been using the K-Fold cross validation a few times now to evaluate performance of some learning algorithms, but I've always been puzzled as to how I should choose the value of K. I've often seen ...
19
votes
6answers
699 views

Having a job in data-mining without a PhD

I've been very interested in data-mining and machine-learning for a while, partly because I majored in that area at school, but also because I am truly much more excited trying to solve problems that ...
3
votes
1answer
49 views

What is a distance or similarity metric that takes into account the improbability of a match?

Suppose I want to measure similarity between users. If two users match on an item that is very improbable, I want to give greater weight to that.
0
votes
1answer
50 views

How do you find the best feature combination to achieve maximum accuracy?

I have a set of 9 possible features that I can extract from an image. Does anyone know of a method that I can employ to utilize the best feature combination to get the maximum possible classification ...
0
votes
0answers
29 views

Creating sample cases from a random forest model in R

Along similar lines to this post: Obtaining knowledge from a random forest . I was wondering if there is an easy way in R to generate some sample cases using the output of a random forest model in R. ...
4
votes
1answer
57 views

Bayes Network/Conditional Probability Visualization Tools

I'm hoping that someone could suggest a tool for viewing conditional probabilities withi. I am currently using Weka, but the ability to view the conditional probability tables of nodes within the ...
2
votes
0answers
50 views

What machine learning algorithms are used for internet advertising?

I.e. based on stats of site visiting, clicks, classifying visitors based on what they are interested in, time, day of week etc. Any frameworks/resources you can recommend? For example: we have a lot ...
1
vote
0answers
48 views

How to compare the accuracy of two different models using statistical significance

I am working on time series prediction. I have two data sets $D1=\{x_1, x_2,....x_n\}$ and $D2=\{x_n+1, x_n+2, x_n+3,...., x_n+k\}$. I have three prediction models: $M1, M2, M3$. All of those model ...
3
votes
0answers
36 views

Minimisation algorithm for a mix of discreet and continuous parameters?

I have a minimisation problem in which the parameters are a mix of integers and scalars. Some of the integers have a small range, around 0-10 but others range in the thousands. To give some context, ...
0
votes
1answer
44 views

How to select validation data when training a neural network?

I am training a neural network with time dependent financial data. In order to avoid overfitting I would like to stop the training at the point where my neural network stops improving on a set of ...
1
vote
1answer
142 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 ...
5
votes
1answer
111 views

Can I have too much data?

I am training a large set of neural networks for a quite simple task. 10 of the networks have the same configuration, but have different amount of data. The 10 networks each have one hidden layer, ...
1
vote
1answer
65 views

Regarding boosting, bagging and bootstrapping [closed]

How to understand the relationships, comparative advantages, and comparative disadvantages of boosting, bootstrapping and bagging in terms of their respective applications in data mining.
0
votes
1answer
43 views

Finding weights for variables in kNN

I'm using euclidean distance for kNN. I have labeled data, I have took logarithm of some variables to make them look more like normaly distributed and scaled them all. And now I would like to multiply ...
4
votes
0answers
68 views

Fast alternatives to the EM algorithm

Are there any speedy alternatives to the EM algorithm for learning models with latent variables (especially pLSA)? I'm okay with sacrificing precision in favor of speed.
1
vote
1answer
41 views

Is there a sequential version of probabilistic latent semantic analysis?

Does someone know if it exists some way to do online learning with pLSA? The model training is really time consuming, so it is not feasible to rebuild it after every changes in the data.
3
votes
1answer
44 views

Validation techniques for hierarchical model

I have a hierarchical model that I need to validate. My model is as follows: we have a collection of $\lambda_i$ that we draw from $Gamma(\alpha,\beta)$. Then, we draw our data point $y_i$ from ...
2
votes
1answer
79 views

Utility or software to visualize Neural Network?

I am using Octave to generate a Neural Network with a single hidden layer, and saving it as two CSV files. Is there a utility or software that will load the files and create an image, PDF or HTML ...
0
votes
0answers
25 views

SVM using RBF and nearest neighbor classification method

SVM using RBF kernel is claimed to be similar (equivalent) to the K nearest neighbor classification method. I am not very clear about the analysis process of building this kind of relationship. Thanks ...
1
vote
1answer
67 views

Ensembling regression models

I'm working on a securities pricing project and have a bunch of models I'd like to stack/ensemble together. I've been using simple linear regression in R (the lm() function) so far but the results ...
-1
votes
0answers
36 views

Feature selection based on univariate test and multivariate test

I have learned that there are feature selection methodologies based on univariate tests and multivariate tests. It looks to me as if they are not covered in the classical machine learning books. Do ...
1
vote
2answers
66 views

Kernel logistic regression

I heard Kernel Logistic Regression is a classical combination of kernel methods and Logistic regression, but I cannot find any major reference (book, or paper) on this topic. Can you give me any ...
0
votes
1answer
49 views

Using GPML in Matlab for MultiClass Classification

I am using Rasmussen's GPML code in Matlab R2011a_student. I have training data (2560x29707) w/ labels (6 classes), and test data (640x29707). To prep the data I have 1. converted from sparse to full, ...
0
votes
0answers
33 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 ...
4
votes
0answers
105 views

First step for big data ($N = 10^{10}$, $p = 2000$)

Suppose you are analyzing a huge data set at the tune of billions of observations per day, where each observation has a couple thousand sparse and possibly redundant numerical and categorial ...
0
votes
0answers
18 views

How to defined a task in multi-task learning

In Multi-task logistic regression, what exactly is meant by a task? I reviewed the following paper: Xue, Ya et al. “Multi-Task Learning for Classification with Dirichlet Process Priors.” The Journal ...
1
vote
0answers
42 views

Software library for Hidden Markov Modeling of a large text database

Given we have a large database of texts (e.g. product descriptions) and we want to extract multiple types of information (e.g. brand, release date, features, price, etc.) what's a good library to ...

1 2 3 4 5 … 12