Questions tagged [classification]

An instance of supervised learning that identifies the category or categories which a new instance of dataset belongs.

Filter by
Sorted by
Tagged with
0
votes
1answer
30 views

Incorrect Text Classification, But Accurate Model. Do I Perform Manual Text Classification For A Data Set?

I'm currently using Google's BERT pre-trained sentiment analysis model that is trained on an IMDb pos/neg review dataset. I'm using this model to predict whether tweets are positive (bullish) or ...
6
votes
1answer
171 views

What makes binary cross entropy a better choice for binary classification than other loss functions?

I'm reading this post where I came across this quote "Cross-entropy is the default loss function to use for binary classification problems." But what about it makes it the default and presumably best ...
0
votes
1answer
18 views

How to interpret a random variable in the variable importance?

I have a problem, for simplicity let's say it is a binary classification problem. I am trying to solve this problem using XGBoost. A standard output plot for any ML algorithm, is the feature ...
-1
votes
2answers
21 views

Is “two-feature classification” appropriate and clear to describe this figure?

it seems that this figure can be used to elaborates the perceptron model and SVM model: Is it appropriate and clear to call this figure "two-feature classification"? Is this a canonical name?
0
votes
1answer
27 views

What is the geometrical representation of the y value in a 2-variable perceptron plot?

This figure represents a perceptron model with a 2 dimensional feature vector input. The hypothesis space of the perceptron is defined by this set: $\{y | y = w\cdot x + b\}$ What is the ...
1
vote
0answers
42 views

Keras Model always predicts the same class

I am currently trying to build a CNN classifier which takes a ector representing the log of an ECG spectrogram together with its class. What I am currently experiencing is the fact the model always ...
1
vote
1answer
22 views

non-separable assignments of the vertices of a hypercube

I have a question regarding exercise 14.17 in An Introduction to Information Retrieval by Manning et al. The problem is: "Assuming two classes, show that the percentage of non-separable ...
3
votes
1answer
23 views

How to Determine Specific Activation Function from keras' .summary()

I'm following a tutorial where a particular model is provided in .h5 format. Of course, I can call model.summary() on this model after loading it with ...
3
votes
1answer
26 views

How to prevent model from recognizing false Classes

Let's say that I have a model that can recognize Cats and Dogs. However, when I use a picture of a Cup or Human it generates a random prediction at 0.70 confidence. Should I use sigmoid instead of ...
0
votes
0answers
34 views

Do I use class weights to penalize false negatives or threshold optimization to improve recall?

I built a Random Forest model for a binary classification problem.Both the classes in the target variable are balanced. My main class of interest is 'class 1'. False negatives are more costly to me, ...
1
vote
2answers
623 views

ValueError: Found input variables with inconsistent numbers of samples [43,19]

So, I've been trying to split my dataset into a 70-30 ratio using train_test_split in order to work things out with sklearn's ...
0
votes
1answer
30 views

What is the intuition behind using LSTM for classification tasks?

LSTM is good for sequence prediction, because it can remember the previous context. What is the rationale behind using it in classification tasks ? In particular, they have used it for the following ...
0
votes
1answer
147 views

No correlation found between dataset features

I'm trying to build a classification model that predicts the price of New York taxi trips (year 2018). Datasource page Since the original file is very large (112 234 626 rows), I constructed the ...
0
votes
1answer
35 views

What could make a set of the train data more predictive than the whole train data

I took a sample of my training data and balanced it and then trained my model. The results obtained are more accurate than using the whole set of train data (balanced or imbalanced). My question is: ...
2
votes
5answers
96 views

Large no of categorical variables with large no of categories

I'm working on a binary classification problem where the dataset is slightly imbalanced (30% class 0 | 70% class 1). Most of my features are categorical with large number of categories. For example: ...
2
votes
1answer
37 views

Multiclass classification with high number of classes, high number of features and small sample size

I am working on a biology related dataset with over 300K features, and I only have about 5K samples. I want my model to classify many classes. For this problem in particular the class is age. Each age ...
3
votes
1answer
67 views

Geometric interpretation of MLP output

I am really interested in the geometric interpretation of perceptron outputs, mainly as a way to better understand what the network is really doing, but I can't seem to find much information on this ...
4
votes
1answer
67 views

Training textblob with 16k rows of labeled data won't work (only few are working)

I've got labeled data in a csv which looks like: title,type Women Jacket A,Clothes Mens Running Shoes B,Shoes Children backpack,Bags and a script: ...
0
votes
0answers
24 views

Python: OneVsRestClassifier results in incorrect hyperplanes with images (linear classification)

Using the OneVsRestClassifier leads to two hyperplanes which have a huge y-axis distance to the actual min. y-value of the data points (somewhere in the area of -300 while the y-minimum is around -1). ...
2
votes
1answer
44 views

What's a classifier capable of predicting a variable number of classes

I want to solve what I understand as a classification problem regarding tagging. Let's say an Entity can have 0 or more tags and I want to be able to predict which tags (if any) an entity might get I ...
1
vote
0answers
19 views

Python: Why does the OutputCodeClassifier provide more estimators than there are classes?

Why is len(clf) = 6 even though there are just 4 classes? Data is stored in the following format: ...
1
vote
1answer
43 views

Unseen classes - 'Other' class or classifying based on the existing classes

Let's suppose that I have a classifier which detects whether a script of code is written in Python or C++ or Java. Therefore, my classifier has 3 outputs which give the probabilities that a certain ...
1
vote
4answers
43 views

Is it OK to train a binary classifier using all the extremely imbalanced data if the majority class is negative?

I'm training a neural network as a binary classifier for text classification. The data is very imbalanced, where the ratio of TRUE:FALSE is approximately 100:10000 Intuitively, it feels like using ...
1
vote
1answer
26 views

Split large documents into many smaller ones?

I want to build a basic language detector for English, French and German. I went to wikipedia and I downloaded the page of 'Technology' in all these languages. In all these cases, we are talking for ...
1
vote
1answer
28 views

Fewer observations & larger documents vs More observations & smaller documents

Let's suppose that I have a dataset of 1000 documents. Each document is a restaurant review (so relatively short text) and it has labels {Negative, Indifferent, Positive}. Let's suppose that the ...
2
votes
1answer
41 views

TF-IDF vs TF for classification

Let's suppose that I have a dataset of 1000 documents. Each document is a restaurant review (so relatively short text) and it has labels {Negative, Indifferent, Positive}. Let's suppose that the ...
3
votes
1answer
50 views

Classify samples based on other sample probabilities

I was wondering if there's a way to train a classifier or set up a way of classifying after that can classify certain samples as some relationship between the previous two. I notice that, for example,...
1
vote
0answers
47 views

Integration of NLP and Angular application

I'm doing a small POC in which I've trained my Machine Learning model (Naive Bayes) and is saved in ".pkl" (pickle) format. Now my next task is to develop a web application which asks the user to ...
1
vote
1answer
28 views

How to handle overfitting in the following classification case

The confusion matrix is as below :- [[ 0 0 5 1 0 0] [ 0 0 19 14 0 0] [ 0 0 217 151 0 0] [ 0 0 84 282 0 0] [ 0 0 6 111 0 0] [ 0 0 0 10 0 0]] ...
0
votes
2answers
41 views

A robust metric in the presence of class imbalance

When evaluating the performance of a multiclass classification problem, on a highly imbalanced dataset, what is the most robust metric for this purpose? I read a paper that states: "Average ...
0
votes
1answer
29 views

Choice of method for classification and/or duration prediction

Let's consider an e-commerce problem. I have data about users that almost place an order online : they give some information about themselves (name, age, etc.), but won't immediately validate the ...
0
votes
0answers
24 views

Open Set Classification : How to count True Positive, False Positive,

TL;DR : How to compute TP,FP,TN,FN in Open Set Classification setting ? Even if the problem is simple; the answer may be tricky, so is my question. Given two sets of clusters : one from experiments, ...
3
votes
1answer
75 views

Hyperopt vs Default Values

When I use the hyperopt library to tune my Random Forest classifier, I get the following results: Hyperopt estimated optimum {'max_depth': 10.0, 'n_estimators': 300.0} However, when I train the ...
1
vote
0answers
18 views

feature extraction for Radio signals classification

I found some code where the developer is trying to solve the problem of "Radio signals modulation classification" in this link link one of the solutions is using SVM to solve the classification ...
2
votes
1answer
113 views

Definition of Type 1 and Type 2 Errors in Classification

Why are the type 1 and type 2 errors as defined in bankruptcy prediction, different from type 1 and type 2 errors based on confusion matrix? In bankruptcy literature: Type 1 error: predicting a ...
1
vote
2answers
42 views

Classification task - how to handle large data file? [closed]

I'm planning to construct a classification model for predicting New York taxi trip fare. The CSV-datafile for this is very large, containing 112 234 626 rows (ca 10 GB). I have managed to download ...
2
votes
1answer
52 views

How to Setup Multivariate Time Series Dataset for Classification

First post on StackExchange. I’m fairly new to ML, with about 1 year of experience so please pardon any ignorance or misuse of terms. I have a multivariate time series dataset where I would like to ...
0
votes
1answer
30 views

Random Forest application with 40+ Predictor Variables

I am using R package randomForest to build a Random Forest model for classification. Ultimately, I need to choose one of five programs for a group of individuals ...
0
votes
0answers
20 views

Variable Importance in sklearn's RandomSearchCV

I ran an algorithm for fitting with RandomSearchCV on a classification problem, the results were good but I need to know which variable had more importance. I know that a fitting without ...
0
votes
1answer
59 views

Best metric in imbalanced classification for multi-label classification

My test data are imbalanced, i tried to use the precision or the gmean as metrics for a multi-label learning model, but both metrics are not very informative. Is there any way to use for example the ...
1
vote
1answer
44 views

Hand-crafted decision tree inspired from learned decision tree

Goal of this question: As I am the only 'machine learning guy' in our group, I wanted to get an outsiders view, that is a sanity check if what I am doing adheres at least to 'decent practices' in ...
0
votes
1answer
35 views

For binary classification, which is best Random Forest or Neural networks?

I had to perform a binary classification, and from the beginning I started thinking about using the Random Forest classifier. But now I'm thinking, if using a neural network would've not been better. ...
1
vote
0answers
19 views

Penalization term for unfairness

I am reading [1], where the researchers do a logistic regression, but add to the loss function the following penalization term for fairness $ R^{AVD}_{FP}(\theta; S) = \left\lvert \dfrac{\sum\limits_{...
0
votes
2answers
35 views

Is the decision tree the right choice to classify for this dataset?

I have a bunch of tuples like this; [SourceIP, DestinationIP, Port, TimeStamp] If a destination IP recorded 21, 22, 23 and 80 port (set of 5 tuples) then I will decide something, if it has set of 4 ...
0
votes
1answer
25 views

Using sigmoid in binary DNN output layer instead of softmax?

For a binary DNN, the output is $y_0 + y_1 = 1$ since they are the probability distribution, hence the sum must equate to 1. However, I've been told that $y_1$ is sufficient to represent the output of ...
0
votes
1answer
19 views

Evaluating performance of classifier on lopsided dataset

I have a binary classifier that I would like to evaluate the performance of. It's been both trained and tested on a data set where the ratio of true to false labels is lopsided. This means that while ...
1
vote
0answers
15 views

Perpendicularity between random variables?

I am reading Bechavod et al. (2017) [1], and at page 3 there is written: In the example, each data point lies in $X = (X_1,X_2) = \{0, 1\}^2$ and has two features—$X_1 = A$ is the protected ...
0
votes
1answer
33 views

How to Interpret the ROC Curve?

i plotted the ROC curve for RandomForest Classifier and this is what i get : The shape looks weird to me , can somebody help me to make sense of it , and is this shape 'common' to not say normal? ...
1
vote
2answers
30 views

How can I measure if a population has the same distribution as other?

Is there a population similarity index of some kind which could help me determine if two populations in two different datasets are the same or at least similar? The datasets have the exact same ...
0
votes
0answers
13 views

Bayesian Statistics — Classification Using Probabilistic Inputs

would like to hear people's opinion on the problem. I am working on a project and would like to know if on right path. Essentially, I have a segment of 100k people from a population i'm trying to ...