Questions tagged [classification]

Statistical classification is the problem of identifying the sub-population to which new observations belong, where the identity of the sub-population is unknown, on the basis of a training set of data containing observations whose sub-population is known. Therefore these classifications will show a variable behavior which can be studied by statistics.

2,260 questions with no upvoted or accepted answers
Filter by
Sorted by
Tagged with
0
votes
0answers
29 views

Training accuracy changes as lot

A i am currently training a phoneme detector using a yes/no data set.. but the training accuracies seem a bit weird.. accuracies changes constantly.. Not quite sure How i should interpret.. Is it ...
0
votes
0answers
38 views

Comparing percentages

Question: I am trying to compare the percentages I have accumulated from a research study. Say that I have collected data on a certain group of patients, all having the same diagnosis. Then I divy ...
0
votes
1answer
82 views

Can one leverage the probability difference between the the predicted class vs the original class?

I have P predictors, C classes After performing training on a training partition I run the test partition through my classifier. Say for a particular test instance the predicted class is Cpred with ...
0
votes
0answers
25 views

ROC curve issue for multiple users

I have a system that has a classifier and multiple users. To test each user I train a binary classifier for each genuine user and all other users. I find that the score scales seem to be different. ...
0
votes
1answer
84 views

Are Linkage Distance, Within groups sum of squares and Minimum Distance within clusters different

I am doing a cluster analysis and confused with three terms. Are Linkage Distance, Within groups sum of squares (total within sum of square), and Minimum Distance within clusters different when you do ...
0
votes
2answers
697 views

Correlated variables - classification

I have some data set and need to use a few classification methods to make prediction. I first need to pre-process the data set. France is administratively divided in regions (13), and regions are ...
0
votes
0answers
72 views

Naive Bayes for GPA classification?

So I am doing a project where I have to find a method that uses a lot of both categorical and numerical variables collected from surveys to predict a child's "discretized" GPA values. For example, ...
0
votes
0answers
106 views

How to penalize coefficient of some features in a model?

I am building a logistic regression model for classification purposes -- I have 8 features - a1, a2, a3, a4, b1, b2, b3, b4. I want to build a model such that I should be able to penalize the ...
0
votes
0answers
728 views

Cross Validation: metrics to determine overfitting?

Let's say I'm doing a multivariate regression and I perform a k-fold cross validation. I'm aware of the CV error term, which is basically the average prediction error across every fold. My question is:...
0
votes
0answers
230 views

tf-idf, many docs, many terms

I have many documents and many labels. I would like to extract those terms that have a high tf-idf among the documents of a particular class. ...
0
votes
0answers
180 views

SVM performance drops in validation

I'm checking my ability to predict a binary outcome using SVM. While the model fits very well, when resampling the model performance drops (but not so much when using logistic regression) which makes ...
0
votes
0answers
82 views

Design of a triangular test (or 2-out-of-5) when there is intra-group variability of samples

I would like to submit to an expert group a set of images from two genetically different types of plants to see if they can find a difference. I have 20 images for each of the two types. Images are ...
0
votes
0answers
567 views

Classifiers for small data sets and low dimensional features?

I have a small data set (under 100 samples) and 5 input features. I often hear about how neural networks are prone to overfitting under such conditions and that naive Bayes is likely to underfit. ...
0
votes
0answers
69 views

Valid to supply unique features to each fold of k-fold Cross Validation?

I have a dataset of n = 55 samples with 300,000 - 600,000 features for each sample. I am trying to train a model (ksvm classification) to predict the class of each sample. N = 26 of the samples are ...
0
votes
1answer
33 views

Binary classifier for cross-related and time-varying points?

Suppose one has $N$ sensors labeled $1,2,3\dots$ arranged in an equally spaced grid. For each point you have a measurement that can vary in time $X^{i}(t), i \in {1,2,3\dots}$. The point $A$ needs ...
0
votes
2answers
142 views

Would it make sense to classify pdf documents based on their structure with RNNs?

I am wondering if using RNNs to classify pdf documents for types of documents (i.e. scientific papers, books, reports, etc) based on the structural components like the text boxes and image boxes and ...
0
votes
1answer
2k views

Partitioning with cross validation

I am new to data analytics having only started exploring the field this week. I have downloaded KNIME and am working with a single dataset to try out different classification algorithms. I am ...
0
votes
0answers
1k views

scikit: cross validation with metric='precomputed' and Nearest Neighbors

My goal is to classify samples based on their dynamic time warping distances with k-nearest-neighbor classification. Therefore I compute a nxn matrix, where ...
0
votes
1answer
74 views

C50Rules - how is confidence calculated? (example needed)

In a C5.0 ruleset model, how is the confidence associated with a prediction calculated? I know the help file says ...the predicted confidence value is the ...
0
votes
2answers
38 views

Divide patients into 2 groups

Study design: pre-test (questionnaire) treatment post-test (same questionnaire as in pre). A higher questionnaire score in the post compared to the pre indicated a successful treatment effect. Like in ...
0
votes
0answers
54 views

How to implement test set in classification Model for Prediction?

I have 100 documents which I have converted to Dataframe with the target variable. In supervised learning we need a target variable or label, Y = f(x) So, I created a training set for with my ...
0
votes
0answers
3k views

Covariance matrix for Linear Discriminant Analysis

I read this: But I'm still not sure if I understand what $\Sigma$ is. Does this mean to take the covariance of the entire data set with all classes mixed in, to take the covariance matrix for one ...
0
votes
0answers
192 views

Is the Bayes optimal classifier well defined?

The Bayes optimal classifier (BOC) is defined as follows. When data $D$ is given, the classifier returns the value $$\text{argmax}_{y\in Y} \sum_{h} P(y\mid h) P(h\mid D)\text{,}$$ where the $Y$ is a ...
0
votes
1answer
46 views

What kind of algorithm can perfom classification / regression on numerically ordered and bucketized values

I have a dataset which is labelled with bucketized numerical values. (0-2, 3-8 ... for example). I can transform these values into their centroid and perform a regression but I think the best would ...
0
votes
0answers
123 views

Measuring curve “closeness” with unequal data ranges

Provided that I have a similar example: where the blue data is my calculated/measured data and my red data is the given groundtruth data. The task is to get the similarity/closeness between the data ...
0
votes
0answers
133 views

De-aggregate data: statistically correct approaches?

Reading some research studies and there are interesting aggregated results. I want to plug it into a neural network (or similar) in order to categorise individual entries into alignment with what the ...
0
votes
0answers
49 views

partial least squares for comparing two classes

I have one data set labeled by two classes either A or B. I did PLS and I extracted the variable importance. ...
0
votes
0answers
68 views

Orange SVM versus LibSVM feature type

We are experiencing significant differences in results when using an SVM classifier in Orange versus LibSVM. The only difference in inputs we can find is that Orange identifies features as continuous ...
0
votes
0answers
105 views

Which classifier should I use for sparse boolean features?

I have training data that is classified into 2 categories: X and Not X Each piece of training or experimental data has a variable number of boolean features. Each piece of data may have ~100 features,...
0
votes
0answers
40 views

Differences between these representations of classification problems in probability terms

Say that $\mathcal{X}$ is the set of observations, and $\mathcal{Y}$ is the set of classification labels. Also say that $X$ is a continuous random variable that takes values in $\mathcal{X}$, and $Y$ ...
0
votes
0answers
100 views

Why using RBF helps?

Let's say we are doing logistic regression for classification. When the features are used directly, it means we are using some characteristics (features) of the object to classify them. But when using ...
0
votes
0answers
223 views

Survival Analysis vs. Classification for High Dimensional Dataset

I started learning about survival analysis only recently, so I am not sure if this question fits in survival analysis. If you think I lack the background to do what I am trying to do, please point me ...
0
votes
0answers
206 views

Cross Validation Split Data to test, train and validation datasets + Discretization

I need an advice which portion of a dataset should be used to calculate cuts for discretization. I use two levels of Cross Validation. One is external to the model creating, but the second is used ...
0
votes
1answer
2k views

How to report cross validation?

When we use cross validation (for example 10 fold) I can obtain 10 sensitivity, specificity and accuracy measures and also 10 ROC curves with the associated AUC. What should I do for reporting my ...
0
votes
0answers
592 views

How to evaluate the model performance under multiclass random forest fitting?

I just use the code to run random forest for multiclass response. I binarize the output to apply the ROC score for model performance. Could someone can advice other methods for the performance ...
0
votes
0answers
210 views

When shouldn't we use Centroid-based classifiers?

I understand that Certroid-based classifiers are better than k-NNs. 13.5.1.2 Centroid-Based Classification Centroid-based classification is a fast alternative to k-nearest neighbor classifiers. ...
0
votes
1answer
118 views

Possible machine learning methodology for constructing a classifier on a microarray dataset with limited sample size

I would like to address a specific machine learning procedure I would like to implement in R with the package caret, which is quite challenging regarding its limitations/possible solutions. In detail, ...
0
votes
0answers
118 views

How to get the area under the curve from multivariate analysis

I have two groups, A and B. They have a number of variables as follows Group, Age, Sex, NumberofSomething, NumberofSomething2 I want to determine which of the ...
0
votes
0answers
260 views

Time series classification

I need to find a specific pattern in a continual time series and then to classify it into one of two groups good, or bad. So, the first step would be some kind of search in the time series, but that ...
0
votes
0answers
592 views

Using probability for ensemble classification

I'm running some ensembles of several algorithms and I'm trying to establish possible rules for the final classification process. Right now if have the class probabilities from 3 models. Its a binary ...
0
votes
0answers
433 views

Random splits of datasets influence on decision tree size

If I have a big dataset and I split it randomly to equal size splits. It means that I choose randomly each record and exclusively put in the each split (bagging). One time I split the big dataset into ...
0
votes
0answers
43 views

identifying low information features with $\chi^2$ distribution

On this site TEXT CLASSIFICATION FOR SENTIMENT ANALYSIS – ELIMINATE LOW INFORMATION FEATURES, low information features are identified by using the ...
0
votes
0answers
710 views

imageNet: ground truth in classification task

I hope someone working with the www.image-net.org dataset reads this question. I am confused on what the ground truth is for the image classification task and how exactly the predictions of an ...
0
votes
0answers
434 views

AdaBoost, how to understand the “weighted class probability estimates” in SAMME.R

SAMME.R is a multi-class classifier of AdaBoost. And I am confused about the weighted class probability estimates in SAMME.R. Here is the algorithm: (1) Initialalize the observation weights $w_i=\...
0
votes
0answers
594 views

How to fit a model to my binary time series data using R?

Problem I have a binary time series with a binary dependent variable and several independent variables. What are some appropriate and easy to understand approaches using R that I haven't thought of? ...
0
votes
1answer
50 views

Which attribute selection method to use in clustering in R

I have a number of time series with around 5000 rows. Each row has 6 attributes and a class assigned. My problem is to determine which attributes are more effective in the classification using R, ...
0
votes
0answers
76 views

Classifying spikes as excitatory or inhibitory based on their shape

I have two classes of spike, namely, excitatory spikes and inhibitory spikes. In the figure you can see their shapes. I want to classify my spikes based on these two shapes. The shape in the figure ...
0
votes
2answers
210 views

ranking neural net models with feature selection

I have a sample with around 2000 observations and 10 variables which im using for classification. I plan on classifying the data with a neural net, but before doing so im using Weka's attribute ...
0
votes
1answer
100 views

Convolutional neural network and Transfer Learning

I am using a ImageNet-trained network to extract features and classify my own images. My images are quite different (microscopic images) from cats and dogs but the features extracted from the ImageNet ...
0
votes
0answers
104 views

Will auto-encoder works for already extracted features

Currently, I am try to use auto-encoders based deep learning to perform some classification task. (In particular, I am a newbie and just used Matlab build-in functions instead of other more powerful ...

1
40 41
42
43 44
46