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.

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

Two ML models use different features. Does knowing the features of one model help improve the accuracy of the other model?

Suppose two firms are operating in the same field (e.g. insurance). If firm 1 knows which features firm 2 is using in their model, can firm 1 improve its model using that information? What if firm 1 ...
0
votes
0answers
10 views

normalize output scores for binary classification

I am handling a binary classification problem on an imbalanced dataset. The goal is to create a system able to predict a label between 1 and 10, where 1 means low probability to be in the positive ...
1
vote
1answer
13 views

How to check if there is a linear relationship for a logistic regression model

From what I understand logistic regression expects that there is a linear relationship between the log odds of the target and the feature. Fourth, logistic regression assumes linearity of ...
1
vote
0answers
18 views

PCA returns duplicated features for different components

I performed (sklearn) PCA on a (1416960,140) pandas DataFrame. The resulting components_ attribute is a matrix where each ...
0
votes
1answer
23 views

What are the disadvantages of models with many parameters to tune?

I modeled with 3 different methods but all showed bad accuracy. So im trying to reason why! one of the issues might be the many parameters that i tuned. So i want to know if there are disadvantages in ...
1
vote
1answer
28 views

Reformulation of logistic regression

I am given the question above and can't seem to get the form that it's asked for. I have tried working it backwards from the goal which gives me: ...
1
vote
0answers
13 views

how to z-normalise time series?

Suppose I have 1000 samples of time series, every one of which has 150 points.(If sample frequency is 150 Hz, then every one of the time series stands for 1s.)What is the correct way to z-normalise ...
0
votes
0answers
4 views

What is application of topic model?

For example, I have a collection of documents,then I use document-word matrix transformed from these documents to fit model lda(document_word_matrix) by ...
1
vote
0answers
7 views

Learning to Rank with binary target variable

Is there a Machine Learning Ranking algorithm that can rank documents for a query using as a training dataset to the algorithm a dataset having: some features of the queries, some features of the ...
0
votes
1answer
18 views

Formula for “true distribution” in machine learning problems

In machine learning we often use the concept of a "true distribution" $p_{\mathrm{true}}$ which is sometimes called "nature's distribution" or "target distribution". I tried to express it analytically ...
1
vote
0answers
24 views

T Test Analysis Question

I’m submitting this post in hopes of receiving feedback regarding a project that I’m working on. I need a sanity check to ensure that there isn’t a blatant flaw in my logic that I’m overlooking. ...
0
votes
0answers
3 views
0
votes
1answer
16 views

Breusch Pagan and White test always 0

I'm working on a machine learning algorithm and trying to evaluate the model based on a guide online. No matter how I change the features of the model, the results of both tests are always 0. Is there ...
1
vote
0answers
14 views

Statistical methods to profile clusters

I am working on a clustering exercise and now I need to generate business insights by profiling the clusters(provide a brief description of each cluster. Eg : Cluster 1 mostly consists of high ...
1
vote
1answer
20 views

Train/test split and leave-one-out

I need to build and evaluate a classifier over 100 examples. There are reasons for which I need to use a leave-one-out cross validation approach, and I have a doubt on how I should proceed. Which, ...
0
votes
0answers
18 views

Tiime series forecasing methods for small sample

I have real time data source that emits numeric values every 5 seconds. I wanted to raise alert whenever, for example the last 5 consecutive values, deviate more than a certain level. As you can see ...
0
votes
0answers
12 views

Regularization for regression vs classification [closed]

I'm following the great course of stanford on ML, I was just wondering why the regularization term is different when using regression or classification. In regression we must add the term (lambda/m)*...
2
votes
1answer
21 views

Is the validation data set used for building the testing model?

Let's assume I split my data into 70% training data, 20% validation data, and 10% testing data. For each hyperparameter I am building a model using the training data and determine the best ...
0
votes
0answers
17 views

Linear Discriminant Analysis have Small Sample Size problem (SSS) is it n<<d

It is said that LDA has a Small Sample Size problem (SSS) This problem arises whenever the number of samples is smaller than the dimensionality of the samples. Is this correct that the ...
0
votes
1answer
18 views

How does correlation between independent variable and dependent variable affect models?

I'm trying to fit Logistic Regression and SVM on a standardized dataset with two classes - 0 and 1. It is a balanced dataset. When I plot the feature importance, I see that for both Logistic ...
2
votes
1answer
23 views

Linear Discriminant Analysis for newbie (What is the meaning of dataset is linear separable?)

What is the meaning of "LDA dataset is linear separable"? "the classes are non-linearly separated" "the features have nonlinear relationships" As I know in maths for linear equation and non-linear ...
0
votes
0answers
24 views

This is from Bishop's Pattern Recognition and Machine Learning Book, section 1.6 on Information Theory [closed]

Consider a set of $N$ identical objects that are to be divided amongst a set of bins such that there are $\mathit n_i$ objects in the $\mathit i^{th}$ bin.The total number of ways of allocating the $N$...
1
vote
1answer
23 views

This is some variational calculus used in Bishop's Pattern Recognition and Machine Learning Book (section 1.5.5) on “Loss function for regression”

The expected regression loss is given as:$$E[L]=\int\int \{y(\mathbf x)-t\}^2 p(\mathbf x,t)d\mathbf xdt$$ To minimise the expected loss,Euler Lagrange equation is used which goes like this in the ...
0
votes
0answers
4 views

How to feed LSTM input correctly?

I have a time series problem with 15 minutes as a timestep.The complete data will be from 2016-09-01 00:00:15 to 2016-12-31 23:45:00. I have 5 variables(v1,v2,v3,v4,v5) in the data frame and I want ...
0
votes
0answers
12 views

Can machine learning be used to determine how/when to scan for information

Questions: I want to know if this problem can/should be solved using machine learning? If I can and I should, what resources or next steps should I take ? Which approach seems to fit ? Problem: ...
0
votes
0answers
3 views

svm best paramaters

I am using python for SVM classification and I am trying to determine the optimal parameters for RBF kernel.I use grid search to determine C and gamma. I have images of 256*256 dimension (65536*33),...
0
votes
1answer
18 views

Overfitting with random forest regressor [closed]

Hello I am using random forest regressor and i got RMSE with train : 0.007 but with the test set it is 0.02 so i am having an over-fitting problem how can i reduce the RMSE of test set to 0.007 too ! ...
0
votes
0answers
4 views

NLP features with nonNLP features (particularly time series features)

Long time statistician, new to nlp here. I have become curious on what is the preferred way to mix nlp features wth non-nlp features. For example, say that I had reason to believe that some sort of ...
2
votes
1answer
37 views

How does multicollinearity affect the eigenvalues of a matrix?

I have been looking into ridge regression as a method to address multicollinearity in data. I am aware that multicollinearity can cause high variance in coefficient estimates. I have seen equations ...
2
votes
0answers
39 views

R alternative to scikit-learn [closed]

As a statics researcher, I've been using R since university and I know it quite well, I also know that it's immediate, but it quickly gets chaotic, and this also happens because of the variety and ...
5
votes
2answers
201 views

What is finite precision arithmetic and how does it affect SVD when computed by computers?

Was reading the paper "DETECTING AND ASSESSING THE PROBLEMS CAUSED BY MULTICOLLINEARITY:A USE OF THE SINGULAR-VALUE DECOMPOSITION" by David Belsley and Virginia Klema. After performing SVD, while ...
0
votes
0answers
16 views

Mean average error normalized by the standard deviation of the target

I'm working on a regression model, aiming at prediction age from structured data. I'm using the mean average error (MAE) as evaluation metric, and want to compare my performances with state-of-the-art ...
0
votes
1answer
16 views

What are these features called (LOG and C3)?

I used a feature extraction code, where two of the features are unknown to me. They work well for my model but I don't know the formal names for them. The first one has the following python ...
1
vote
1answer
16 views

What is neural network good accuracy

I am very new at machine learning, and I'm building an artificial neural network that aims to classify inputs into 2 labels. I am training the network with randomly initialized weights and through the ...
0
votes
0answers
19 views

Neural Net - trying to predict that 5 + 5 = 10 [migrated]

I'm learning about Neural Networks and I recently had this idea: trying to give a NN training data of the function $f(x) = 2x$. The question is, can the NN accurately predict that it has to double the ...
0
votes
0answers
4 views

How to customize activation and loss functions for multilabel classification problem?

I'm trying to develop a model using keras able to perform a particular multilabel problem: My target vectors are five components vector in which there are elements between 0 and 4 and Nan. ...
0
votes
2answers
51 views

Am I overfitting even though my model performs well on the test set?

I have a dataset with 1289 observations and around 2000 features. I split my dataset into a 70/30 training and test set. I use GridSearchCV from scikit-learn to perform 5 fold cross validation on the ...
0
votes
0answers
20 views

Best Resource to learn about internal working of ML algorithms for an absolute beginner? [duplicate]

My previous question was about looking for pseudocodes for Boosting algorithms like XGB, Random Forests and LGBM. I figured it would be better if i had some resources to refer to, which would detail ...
1
vote
0answers
7 views

Positional features and feedback loops in ranking

I read the following in the Google "rules for machine learning": Rule #36: Avoid feedback loops with positional features. The position of content dramatically affects how likely the user is ...
1
vote
2answers
16 views

What is the advantage of shuffling data in train-test split?

I have been taking an online course on data science, and was recently introduced the ideas of overfitting and underfitting, that is splitting the dataset we have into two parts into training(80%-90%) ...
0
votes
1answer
17 views

Does having a training and testing data set really work?

Ok, i was just introduced to the ideas of overfitting and underfitting, and its method of detection, that is splitting the dataset we have into two parts into training(80%-90%) and testing(10%-20%) ...
0
votes
0answers
9 views

Statsmodels multinomial logit returning nan for sparse features

I have a travel survey and wish to compare a discrete choice model with some machine learning approaches. For the former, I am using statsmodels MNLogit to calculate the coefficents and p-values of my ...
0
votes
0answers
18 views

How can I add in the functionality “Not found” in k-nearest neighbor?

I'm building onto a image classification library called jFaces. It's made in 100% Java. It don't use Deep neural networks. Instead, it uses PCA and LDA together. Yes, it works. https://github.com/...
0
votes
0answers
88 views

Calinski-Harabasz Index -K-medoids [closed]

Just wondering , can I use the Calinski-Harabasz index to select the number of clusters when using k-medoids for clustering in R
1
vote
1answer
28 views

Re-building a cross-validated SVM

Suppose we are cross-validating parameters of a Gaussian (radial) SVM on $k$ training observations. The parameters are the cost parameter $C$, and the deviation parameter $\gamma$. Then, $4k$ more ...
0
votes
0answers
22 views

How to represent the bias node in simple MLP?

I have a simple 2-2-1 fully connected network, which I suppose means it does not include bias nodes since you don't connect bias nodes at every layer. I have two sets of weights for each layer ...
0
votes
0answers
14 views

Use synthetic data to learn the supervised ML model for real world data

I want to use a large dataset of synthetic data to learn supervised ML model for real world data. For example, I have a large synthetic dataset(these are three-dimensional models) of furniture images ...
0
votes
0answers
4 views

Machine learning technique for text document transformation?

I have a bunch of text documents, split into source documents and transformed documents. These text documents have multiple lines and are edited at specific locations, in a specific way. I make use ...
0
votes
0answers
7 views

How to perform data scaling/standardization on dataset containing grouped values?

So I have a dataset containing the results of executing problem instances with different given solver strategies. Simplified example: ...
0
votes
0answers
2 views

Data-preprocessing for a machine learning model

I am confused about how to preprocess range based category such as age, tumor-size & inv-nodes. Should I take an average of the limits, as in - 14.5, 24.5 and so on or do one hot encoding of the ...

1
2 3 4 5
298