Stack Exchange Network

Stack Exchange network consists of 175 Q&A communities including Stack Overflow, the largest, most trusted online community for developers to learn, share their knowledge, and build their careers.

Visit Stack Exchange

Questions tagged [machine-learning]

Machine learning algorithms build a model of the training data. The term "machine learning" is vaguely defined; it includes what is also called statistical learning, reinforcement learning, unsupervised learning, etc. ALWAYS ADD A MORE SPECIFIC TAG.

0
votes
0answers
11 views

How can you contribute to solve social problems in Bangladesh by using Machine-Learning Technology? [on hold]

How can you contribute to solve social problems in Bangladesh by using Machine-Learning Technology?
0
votes
0answers
7 views

Type of bias where useless data can discriminate

Let's say that there is a dataset $D$ of samples with domain $\mathcal{C}$ $\times$ $\mathcal{L}$, where $\mathcal{C}$ is a feature space and $\mathcal{L}$ is a binary label space ($0$ and $1$). ...
0
votes
0answers
8 views

Sampling Methods : Bishop

I am reading chapter 11 . Sampling Methods from the book : Pattern Recognition and Machine Learning by Bishop : In the introduction , in short,he evaluates expectation of some function $f(z)$ with ...
0
votes
0answers
14 views

Model Not Performing Well On Validation Data - Customer Attrition Modeling

I am modeling customer churn for the online subscription. I looked back 90 days into customers’ data, using number customer watching behavior etc. I get a pretty strong model based on test data. <...
0
votes
1answer
21 views

Cross Validation Vs Train Validation Test

I have a doubt regarding the cross validation approach and train-validation-test approach. I was told that I can split a dataset into 3 parts: Train: we train the model. Validation: we validate and ...
1
vote
0answers
7 views

Predictive Distribution in Gaussian Process Derivation

In Gaussian Process for Machine Learning (Rasmussen and Williams), on p.11, we are given the following predictive distribution: $$p\left(f_{*} | \mathbf{x}_{*}, X, \mathbf{y}\right)=\int p\left(f_{*} ...
0
votes
0answers
7 views

Empirical Comparison: which ideal data characteristics are best captured by each type of machine learning model?

I have reached the point as a data scientist where the empirical differences between the different types of regression models (leaving out classification only for simplicity) have started to matter ...
0
votes
0answers
20 views

Random Forest Classifier [on hold]

Hi guys I am using Random Forest, SVM, GradientBoosting and Logistic Regression for predicting the Customer Behaviour since when i use these models i always get a 1.0 accuracy but it runs well in the ...
2
votes
1answer
29 views

Why is random sampling a non-differentiable operation?

This answer states that we cannot back-propagate through a random node. So, in the case of VAEs, you have the reparametrisation trick, which shifts the source of randomness to another variable ...
0
votes
0answers
16 views

Gaussian process - Why adding data points cannot increase the predictive bias?

I've seen this question here: How to increase variance in Gaussian Process regression? But a proof isn't provided. I'm looking at this book: Rasmussen Williams 2006 Gaussian processes for machine ...
4
votes
1answer
37 views

Classifier which minimizes inaccuracy

I recently interviewed for a machine learning job which involved very mathematically rigorous questions. This is one of them, which I'm still very confused about. Question: Given a data generating ...
0
votes
0answers
13 views

what is the difference between a multilayered autoencoder and a hierarchical latent variable model?

I have been trying to understand how hierarchical latent variable models are different from multilayered autoencoders and in specific the argument below Autoencoder networks resemble in many ways ...
0
votes
0answers
20 views

Statistically evaluating classification accuracy of machine learning model

Let's say I'm trying to evaluate a classification algorithm and suppose there are $m$ data points in my test set. Here's my understanding so far: assuming my evaluation metric is the classification ...
0
votes
0answers
3 views

Selection smoothing parameter for using function SPM(library SemiPar)

Suppose i fit a model of the following form: ...
0
votes
0answers
19 views

How to compare Accuracy of two RandomForest models? (Chi Sqr or Cohen's H?)

I've got two dataset which have exactly the same structure (15 features, 1 class variable with 7 categories) and roughly the same amount of observations). I trained a Random Forest with the full ...
0
votes
0answers
33 views

Is it possible that PCA works better without data scaling? [duplicate]

I am a beginner. I have a dataset of 1700 samples with 4 features and I have to perform Hierarchical Clustering (the agglomerative version) and I need to decide whether or not to scale the data and ...
0
votes
0answers
9 views

SVM in the classification layer of a Feedforward neural network

I want to use SVM in the classification layer of a 2 layer feedforward neural network. Need guidance from the community on how to approach this problem. This involves capturing the features from the ...
0
votes
1answer
12 views

How to interpret a high sensitivity and low specificity using svm classifier?

I want to have your opinion on the performance of my SVM classifier (k-fold cross validated): Classification of class1(n=45)/class2(n=86) Accuracy: 65.4% Sensitivity: 88.2% Specificity: 22.2% AUC:...
0
votes
0answers
6 views

Scoring the difference between a family of distributions and a test distribution

Let's suppose we have a random model that I can sample to generate distributions of a certain 1D variable. I want to score the distance of a test distribution to the model in question. The distance ...
1
vote
0answers
37 views

Why using RMSE as loss function in logistic regression takes non convex form but doesn't in linear regression? [duplicate]

I am taking this deep learning course from Andrew NG. In the 3rd lecture of 2nd week of the first course, he mentions that we can use RMSE for logistic regression as well but it will take a nonconvex ...
-1
votes
0answers
11 views

How to add regularization term lambda? [on hold]

I have this equation in online-sequential extreme learning machine. how can include a multiplication regularization term lambda to govern how rapidly the network weights update please any help. ...
0
votes
0answers
11 views

Implement the probability distribution function of a dataset and calculate the Mahalanobis distance of any given new instance [duplicate]

I'm looking for some practical guidance, as I have a limited time to go from theory to implementation. I would appreciate concrete recommendation on libraries, tutorials or similar to implemente a ...
0
votes
0answers
6 views

Method to identify recurring patterns (motifs) in a time series after giving a reference pattern

I would like to know if there exists an algorithm using which I would be able to extract repeating patterns from a time series dataset, provided I give it a reference shape. I have included an image ...
0
votes
0answers
11 views

synthesized data to train classifier

Our dataset is relatively small (303 x 14) and so we decided to use synthpop package in R. The basic idea of synthetic data is to replace some or all of the observed values by sampling from ...
0
votes
0answers
11 views

Random effects model vs Region by region model - which is more accurate for translating to business action?

I have 100 regions and several features like [quality_rating, the density of stores in a region, the number of people who purchase, age, gender, income, lifestyle], etc. for each region. Let's say my ...
1
vote
1answer
11 views

Analysis on cluster change

I have 50 datasets each for every year from 1961 to 2010. These datasets keep data about GDP, mortality, natality, etc. My intent is to apply clustering for each dataset and then compare clusters. ...
0
votes
0answers
8 views

What is the scale of pseudo R squared obtained from my Logit model?

I am getting a pseudo R squared in the range of 0.01 - 0.05 when I experiment with various combination of features. I am aware of this post: McFadden's Pseudo-R2 Interpretation says 0.2-0.4 ...
1
vote
1answer
17 views

Including the outcome variable in multiple imputation

I'm trying to perform binary classification on a dataset with missing values. I used sklearn's iterative imputer to impute these values and I got pretty good results. However, I realized that I was ...
0
votes
0answers
14 views

Machine learning for discrete probability distributions when events can be rare

I am following a paper which suggests using a restricted Boltzmann machine for learning a discrete probability distribution. I have encountered a problem, however, when the scientifically "interesting"...
0
votes
0answers
25 views

Can bagging outperform a random forest?

I want to predict wages and constructed bagged regression tree's and a random forest of regression Tree's. The bagged regression tree's outperform the random forest. Is this result even possible , I ...
0
votes
0answers
7 views

Oversampling in Uplift Modelling

I hope any of you can help me in the following matter: I am about to write my master thesis addressing the question how response and uplift modelling differ in terms of performance but also the ...
0
votes
0answers
20 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
12 views

How to interpret precision and recall for multiclass prediction?

I have a few models doing prediction with 4 classes, with the output precision and recall varying with different labels. For example I have (with the class labels being 0, 1, 2, 3 on the x axis): I ...
1
vote
0answers
22 views

Proof for asymptotic error in logistic regression

Ng, A.Y., and Jordan, M.I. (2001). On Discriminative vs. Generative classifiers: A comparison of logistic regression and naive Bayes. Advances in Neural Information Processing Systems, 14, pp. 841-8, ...
0
votes
1answer
16 views

Machine Learning problem - label over multiple lines

Currently I'm trying to work out a project where I would like to recognize movements from videos using machine learning and python. What I've done so far is extracting the x and y values of body ...
0
votes
0answers
9 views

How to calculate accuracy of model without having actual labels [on hold]

i have a dataset having traing.csv and testing.csv files. i have trained the model on traing data and then predict the labels for testing data. as the labels are only given in training data. there are ...
0
votes
0answers
17 views

Unsupervised Learning: Distinguishing 10 spoken numbers in Arabic

I want to develop an unsupervised learning method to identify spoken numbers in Arabic. My dataset consists of MFCC arrays. Every row consists of an array of shape(41,13), The row consists of float ...
0
votes
0answers
15 views

On Structures in Dataset [on hold]

I am working on a dataset challenge and am being asked to detect structures in the dataset. What are some ways we can define structures within the data? Is that pretty much any patterns found? Maybe ...
0
votes
0answers
21 views

Are feature importances from tree based models directly actionable for business?

If my response variable say is "has_repurchased" [0 or 1] and I have all customer level features. Can I rank the features in order of importance from the random forest model and report them as whats ...
0
votes
0answers
11 views

Modeling both linear and non-linear relationship

Apology for being verbose and all the typos or mistakes. This problem has been bothering me for a while and I really hope you can help me with it. Let's say I want to model quarterly sales for a ...
5
votes
0answers
37 views

Asymmetric or unequal misclassification costs in random forest

I have a general question about asymmetric costs. In machine learning problems, there are times when the cost of a false positive is different from the cost of a false negative. Accordingly, models ...
3
votes
1answer
68 views

Features that are important according to random forest are not significant when logit model was used. How to interpret?

I have a feature set for each customer [age, gender, income, lifestyle, & so on...] and a response variable say: has_repurchased. I use a logit model summary which shows income & gender to ...
0
votes
0answers
26 views

How to derive a ranking function from observations

I have a employee dataset with the following 5 details. projects completed customer ratings number of bugs reported customer complaints profit I want to rank ...
0
votes
2answers
19 views

Question from Machine Learning Textbook on Number of Models

In the James, Witten, et.al Statistical Learning textbook, it says the following: "Unfortunately, there are a total of $2^p$ models that contain subsets of $p$ variables." Can someone please ...
0
votes
0answers
14 views

High accuracy on both training and validation but very low on test set

My CNN model has about 96~97% accuracy on both training and validation sets. But when submitting the test set it got only 24% accuracy. Here's my model: ...
1
vote
0answers
22 views

How does regularized regression overcome the p > n problem?

So, I understand why simple linear or logistic regression will have infinite solutions in this case (good answers here and here). But while LASSO will only select n features, Elastic net does not have ...
1
vote
0answers
28 views

K-Fold Cross Validation: Bias in cross-validated effect size measures?

As pointed out by various authors (e.g., Hastie, 2011), K-fold cross-validation has an upward bias of prediction error. I wonder whether the same holds for cross-validated effect size measures such as ...
1
vote
0answers
29 views

Recognizing whether a written and spoken number is the same

For our ML assignment we have three datasets. The challenge is about checking whether a written and spoken number refer to the same number. We're using the MNIST dataset with handwritten numbers, and ...
0
votes
0answers
16 views

What are other nonlinear transformation methods in machine learning except Neural Network activation functions?

One advantage of the MLP neural networks is the nonlinear transformation used on raw features. The popular ones used are the activation functions like Sigmoid, Tanh, ReLU, Leaky ReLU, etc. They are of ...
1
vote
1answer
27 views

What units is my mean squared error if I center and scale my training data?

I have a KNN model that I used to predict the close price on houses. ...