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.

Filter by
Sorted by
Tagged with
0 votes
0 answers
6 views

Expert-model and machine learning hybrid approach

Is there a recommended approach how the results of an rule-based expert system can be combined with an ML model? Let's say you have a text classification problem with 100 classes. For some classes you ...
0 votes
1 answer
356 views

XGBoost, Imbalanced Data and CalibratedClassifierCV

I am currently working with a slightly imbalanced dataset (9% positive outcome) and am using XGBoost to train a predictive model. ...
1 vote
0 answers
16 views

How to deal with unknown classes with a convolution neural network classifier?

I'm quite new into the DL and ML field. I'm training a CNN able to classify 3 different classes, however I would like in the testing phase to make the CNN able to not misclassify images that do not ...
2 votes
1 answer
2k views

How to manually balance unbalanced multi-class/multi-label data?

I have a multi-class and multi-label classification problem, i.e.: each sample can have more than one label associated to it and there is a total number of M ...
3 votes
1 answer
59 views

Usefulness of KS tests and other similar distribution comparing tests

I am working on a machine learning binary classification problem. I have an outcome variable status called as loan paid and <...
0 votes
0 answers
10 views

Post analysis using raw data or SHAP values in Machine learning

Let's say I have SHAP value returned in dataframe for input variables like below ...
8 votes
1 answer
2k views

When is oversampling poor practice?

For my particular domain and problem, I have data on the entire population. However, my "event" only occurs in 0.5% of the cases. I want my model to be able to pick up on significant characteristics ...
0 votes
0 answers
9 views

Why PR score is down when balanced accuracy is good?

I just read this discussion here and here. I have a dataset of 977 records where class proportion is 77:23. My balanced accuracy is 75.5, ...
0 votes
0 answers
8 views

What methods can I used to assign cases to groups when some variables are dependant on others?

I have a sample size of 42 cases, with about 5 variables for each case. Most of these variables are measurements, and I want to assign the cases into 2 groups (condition present, and condition absent) ...
1 vote
1 answer
217 views

Is it ok to keep a very strong predictor and other weak predictors in the model? The model built is GBM

Age is coming out as a really strong predictor compared to other variables. This is a classification problem, the dependent variable is a (0/1)
1 vote
1 answer
282 views

Assessing overfitting via learning curves

I have intended to find overfitting or underfitting cases. I have used MLP classifier and Logistic regression of scikit-learn. How do I know which is a good fit? Or Which one underfitting or ...
1 vote
0 answers
12 views

Optimize classification rule in multinomial logistic regression

We know that in the case of logistic regression, a classification threshold p=0.5 is generally not an optimal choice when seeking to optimise sensitivity and sensitivity. This is generally due to the ...
0 votes
0 answers
8 views

Why variable representation plays a role in prediction?

I am working on binary classification using a random forest, where the data have 977 records and 6 columns. The class ratio is 77:23. I have two derived input variables. One variable is called ...
0 votes
0 answers
8 views

Precision vs Recall Tradeoff plots 2 separated lines

I'm trying to build a binary classifier with high recall and slightly better precision so as to avoid a lot of False Positives. So far the best scores I have got from all different types of model ...
1 vote
2 answers
67 views

Is it normal for simple logistic regression to significantly outperform any other statistical ML algorithm?

I'm working on a simple classification project with an imbalanced (minority-to-majority-ratio ~ 0.2) dataset that has ~4000 rows and ~200 features. I noticed that, for my dataset, a simple logistic ...
1 vote
1 answer
2k views

LibSVM - Multi class classification with unbalanced data

I tried to play with libsvm and 3D descriptors in order to perform object recognition. So far I have 7 categories of objects and for each category I have its number of objects (and its pourcentage) : ...
3 votes
1 answer
730 views

How to build the feature vector from sentence for intent classification in NLU?

I am trying to develop a NLU (natural language understanding) engine which interprets human language utterance to intent and slots. After some searching, I found this very useful question for NLU ...
0 votes
1 answer
491 views

Almost all predictions of a SVM are positives(or are all negatives)

I'm facing a binary classification problem using svm light. However using 5-fold-validation I noticed that later I train SVM with training set (Half positive and half negative samples about) the ...
0 votes
0 answers
15 views

How to conduct t-test for comparing the accuracy of two binary classifiers? [closed]

I am using two binary classifiers that predict the accuracy of samples over a dataset. I need to check if the difference in the mean accuracy between the two models is statistically significant. ...
3 votes
1 answer
15k views

how to prevent overfitting with knn [duplicate]

Using too low a value of K gives over fitting. But how is overfitting prevented: How do we make sure K is not too low And are there any other precautions taken in k-nn that help prevent over ...
1 vote
0 answers
10 views

How do I perform a train-validation split on data with class imbalance such that the class imbalance ratio is preserved?

My data has class imbalance-- that is, some classes have significantly fewer training samples than the others. I want to perform a train-validation split in such as way that the class ratios are ...
0 votes
1 answer
183 views

multi-class (granular and coarse) classification

I'm trying to solve an image classification problem that involves a very large number of classes. Each image actually has two labels associated with it that we can think of as a coarse and a granular ...
1 vote
0 answers
13 views

Risk score uncertainty quantification

I am working on various risk score estimation problems. I assume individual subjects are associated with a true risk $$ r_i = f(x_i) + \varepsilon_i$$ where $x_i$ is some available information about ...
2 votes
3 answers
322 views

Tensor Classification Models

Aside from Convolution Neural Networks, are there any other methods that allow for classification of Tensors? My observations consist of multi-dimensional tensors with height of 1, where each channel ...
3 votes
1 answer
156 views

If the AUC score is 100 percent can F1 value be 99.94 percent?

If the AUC score is 100 percent can the F1 value be 99.94 percent? I would expect 100 percent, too.
1 vote
1 answer
10 views

High Performance Classification or Similarity Algorithim for Mixed Data Types?

I have a database holding 10-ish features that describe different breeds of dogs. They are mostly categorical features, but some provide ranges for values. Here's a demo representation of the database,...
0 votes
1 answer
170 views

What are some alternatives to Hidden Markov Models for Part of Speech tagging?

I wanted to build a POS tagger but found the HMM Tagger to be too mainstream. Any other Classifier/Statistical Model that I could use?
1 vote
1 answer
338 views

WoE for Random Forest and SVM

There are a lot written about WoE (Weight of Evidence) transformation for the case of Logistic Regression Classifier. It works great. The question: can one (or does it make sense) to use this WoE ...
2 votes
1 answer
488 views

how to avoid overfittig with xgboost and how to increase accuracy

I am doing a binary classification problem, I got to train 85% accuracy, but test accuracy is 72%, I tried different parameters, Cross valid, But overfitting doesn't change, please help me on how to ...
0 votes
1 answer
19 views

Churn model- how to handle new users without enough historic data?

I'm making a churn model. My observation window (historic data) length is 3 weeks. There are some users that are not been registered to the app that I'm analyzing for three weeks, and as a result, I ...
0 votes
0 answers
12 views

Handle User Behavior Change when creating a ML classifer

I'm creating a churn model. My first thought was that the bigger the training set, it would be better. However, 2020 was a crazy year because the COVID 19. For example, a user who was sick and ...
0 votes
0 answers
21 views

Can I convert a classification problem with an ordinal dependent variable in a regression problem?

It looks interesting to me to know about the variables related to the students performance, so I started to look into the following dataset: https://archive.ics.uci.edu/ml/datasets/Higher+Education+...
2 votes
1 answer
59 views

Clustering while knowing the ground truth: Why would someone choose this approach?

If the ground truth of the class/cluster/segment that our observations belong to, is known in advance, why would someone choose to perform clustering instead of classification? In fact, doesn't the ...
0 votes
0 answers
11 views

Is there a dataset of images with varying sizes? [closed]

I'm working on a project dealing with image classification where I deal with images with varying sizes. I would like to validate my approach in other datasets. I would like to find other datasets with ...
2 votes
1 answer
2k views

Understanding Precision and Recall Results on a Binary Classifier

I know the difference between Precision and Recall metrics in Machine Learning. One optimizes on False Positives and other on False Negative. In Statistics it is called as optimizing on Type I or Type ...
2 votes
1 answer
161 views

Is multiple stage binary classification a good idea if you have very few positives?

The problem is the following: We have a set of, say 5000 documents, with a single binary label. Say that 4900 documents are negative and only 100 are positive. I built a binary classifier while ...
0 votes
2 answers
29 views

Comparing impact of training data size - what testing data size?

I am training a classifier using BERT and want to check how the accuracy changes with increasing training data size. Up until now, I have 1k annotated training samples and tested the accuracy for ...
5 votes
1 answer
576 views

A statistical test to measure the importance of features?

I'm currently trying to assess importance of the features for my classifier. The situation is the following: first I train my classifier with all of the features I have and tested on a test set . Then ...
1 vote
1 answer
255 views

Modifying training set to improve model performance

I am using Logistic Regression (LR) to obtain Coronary Artery Disease CAD probability equation. The dataset has 16 candidate predictors, all continuous. There are two groups, CAD patient group (70 ...
4 votes
2 answers
110 views

Why exactly does a classifier need the same prevalence in the train and test sets?

Here are some commonly seen statements about the importance of prevalence in the train and test sets when developing a classifier: "Another reason not to rebalance datasets is that models should ...
0 votes
0 answers
13 views

Possible ways to feed two variables containing text data into an ML model in an NLP problem

In a natural language processing (NLP) problem, we have a couple of variables, say A and B. A denotes a phrase (1-2 words) and B denotes another phrase (>3 words). There is one target variable that ...
2 votes
2 answers
303 views

class weighted classification

I am working on my multi-class classification project and I have a question: I have three classes in proportion: 50%, 47% and 3%. I decided to use class_weight="balanced" parameter in random ...
0 votes
0 answers
10 views

GPT-2: Why should text classification work on the last output embedding?

When GPT-2 is fine-tuned for text classification (positive vs. negative), the head of the model is a linear layer that takes the LAST output embedding and outputs 2 class logits. I still can't grasp ...
2 votes
1 answer
22 views

Will a dataset with multiple labels perform better than with binary labels?

Suppose I have a dataset comprised of garbages. Will a model perform better if I only label the dataset with biodegradable or non-biodegradable? Or will it be better if I label them with plastics, ...
2 votes
1 answer
866 views

Semantic Segmentation Multi-Class Single Channel Output Math

For semantic segmentation problems, I understand that it's a pixel-wise classification problem. At the last layer of the neural network, I would basically have a 1x1x1 convolution layer with a softmax ...
0 votes
1 answer
621 views

unsupervised Time series anomaly detection

I have 3D printer that working exactly 400 second for printing element X [0-400]. The printer produce 30 signals (features like VOLT,X,Y,Z,TEMP etc') in frequency of 50HZ (every sample 0.02 ms) ,for ...
0 votes
0 answers
8 views

Windowing timeseries classification data

INTRODUCTION: So basically, I have a dataset with 6 columns and around 10k rows. The output column is a label corresponding to every row, with the labels being 0 and 1. The dataset is timeseries based....
1 vote
1 answer
36 views

Why is a random forest regressor better than a random forest classifier when predicting a category?

I am building a model that recommends the optimal golf club based on data I have gathered. Since the model prediction should be a category, ie. a golf club, I would assume I would have to use a ...
0 votes
0 answers
7 views

Fuzzy membership - Kernel SVM in R

I'm trying to perform a binary classification task using SVM with radial basis kernel in R and I want to assign fuzzy memberships to the datapoints. Already existing function in R packages such as ...
3 votes
2 answers
184 views

AUROC too high in image classification

I'm dealing with an image classification problem, with a multiclass imbalanced dataset (the bigger class has 4000 samples and the smaller has 110 samples) with 50 classes and 24000 samples. I'm using ...

1
2 3 4 5
129