Questions tagged [machine-learning]

Machine Learning is a subfield of computer science that draws on elements from algorithmic analysis, computational statistics, mathematics, optimization, etc. It is mainly concerned with the use of data to construct models that have high predictive/forecasting ability. Topics include modeling building, applications, theory, etc.

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

Using Sentence-Bert with other features in scikit-learn

I have a dataset where one feature is text and 4 more features. Sentence-Bert vectorizer transforms text data into tensors. I can use these sparse matrices directly with a machine learning classifier. ...
2
votes
0answers
13 views

Random Forest importances vs Feature Permutation: cummulative sum of importances are 1 and 0.1, respectively. Make sense?

I am performing feature selection by using two methods: MDI (RandomForest importances) and Feature Permutation, in order to compare what are the features considered relevant for both methods. My ...
0
votes
0answers
9 views

BayesianOptimization tuning the same parameter with different results

I'm running a hyperparameter search using Keras wherein there is only one hyperparameter explicitly specified (# of LSTM units). However when running BayesianOptimization, after a while I notice it ...
0
votes
3answers
26 views

I come across with this code. I would like to know what it means in R Programming

A training dataset, which contains 150 columns, then I found the following code somewhere, training <- training[, 6:dim(training)[2]] what I understood is it is ...
0
votes
0answers
12 views

Random Forest to Identify Page: Feature Selection

I am new to machine learning. I know of a project that used Random Forest to identify the type of pages in financial reports, e.g., identify if a page is the cash flow or income statement. The ...
1
vote
1answer
14 views

How do you select the appropriate features and plot the data it so that a k-means algorithm can cluster it?

I am still just dabbling in the shallower waters of machine learning and I am looking to compare the results of a Supervised algorithm (KNN) and Unsupervised algorithm (k-means) when it comes to ...
1
vote
0answers
12 views

Clustering with hierarchical data dependencies

I am currently looking how to cluster data with hierarchical dependencies. An example of a problem that I want to cluster: we would to cluster cities to identify similar characteristics with respect ...
1
vote
1answer
27 views

What's the correct cost function for Linear Regression

As we all know the cost function for linear regression is: Where as when we use Ridge Regression we simply add lambda*slope**2 but there I always seee the below as cost function of linear Regression ...
0
votes
0answers
12 views

What effect repetitive data will have on the performance of the model

I understand that my question is very broad and that the correct answer may depend on various things. I want to get an idea in general what we may expect if we have repetitive data in our dataset. ...
0
votes
1answer
11 views

How can I learn to better explain architectural choices?

I've found out that most of the choices made during model selection are based on a sort of trial and error. From what I've heard, even the most experienced Data Scientists cannot know beforehand ...
1
vote
0answers
11 views

What AI predictions can be done using Temperature and Current features of electrical assets?

I have recently deployed hundreds of IoT based Temperature/Humidity and 3-phase Current sensors on electrical assets/machines(e.g. welding machine, integrated panel etc) in a warehouse and waiting for ...
0
votes
1answer
16 views

If we train a model every time from scratch by using current task and samples from memory (ER) then is it correct way to perform continual learning?

Suppose that there are T tasks. We use an experience replay (ER) strategy using a tiny episodic memory. Here, we train a model always from scratch at each task using current task samples and samples ...
1
vote
0answers
18 views

Predict future payments using Machine Learning

Considering a data set looking like the following, where the values correspond to the amount of money paid. ...
1
vote
0answers
17 views

Does the appliance of R-squared to non-linear models depends on how we calculate it?

Does the appliance of R-squared to non-linear models depends on how we calculate it? $R^2 = \frac{SS_{exp}}{SS_{tot}}$ is going to be an inadequate measure for non-linear models since an increase of $...
1
vote
1answer
30 views

Improving text classification & labeling in imbalanced dataset

I am trying to classify text titles (NLP) in categories. Let us say I have 6K titles that should fall into four categories. My questions: I do not understand why in some ML techniques categories are ...
0
votes
2answers
37 views

Which machine learning model is best for a combination of numerical and categorical data?

I want to develop a ML model which will allow my company to highlight employees which are at a risk of leaving the business, based on a variety of parameters such as performance, absence rates, ...
0
votes
0answers
18 views

Predict value from array (Assisted Sequence Prediction)

TL; DR How to develop a machine learning implementation (in python) that given an array of values is able to predict its maximum percentage of increase? [11.082, 11.052, 11.052, ... 11.202, 11.202, 11....
0
votes
0answers
7 views

Custom vocabulary with Transformer model in Trax

I'm currently working on a programming language translation problem (NMT) using a transformer model with Trax. So I basically need to convert a language A to a language B. In order to tokenize the ...
0
votes
0answers
26 views

Does having two different models improve performance for underrepresented classes?

I am currently working on a dataset that has approximately 7000 annotations, but suffers from severe class imbalance (there are 1331 annotations for the most represented class, and 77 for the least ...
0
votes
1answer
21 views

Why scikit-learn's sequential feature selection requires how much features to be selected beforehand?

From the version 0.24, the scikit-learn has new method 'SequentialFeatureSelector', which adds (forward selection) or removes (backward selection) features to form a feature subset in a greedy fashion....
1
vote
1answer
41 views

How to improve the result of f1 on imbalanced dataset

I have a dataset in which these are the distribution of the data: Neutral. 15000 Negative 3000 positive 2000 And I am mostly interested to improve the ...
0
votes
0answers
14 views

How to evaluate an aspect based sentiment analysis model?

I have a model X that takes a review and predict the position of the aspect and the polarity: ...
1
vote
1answer
22 views

Robustness vs Generalization

I don't quite understand the difference between robustness and generalisability in relation to image processing (CNN). If my model generalises well, it is also robust to changes in the image material. ...
1
vote
0answers
11 views

Methods for comparing datasets for abnormalities

My team at work receives datasets from various companies each quarter, and we're wanting to build some sort of machine learning model that compares the new dataset to our existing data for any ...
0
votes
1answer
20 views

Unsupervised Clustering using K-Modes

I have a problem statement to find the effective time and day to send emails to the doctors. We have doctors belonging to a specific hospital ....also we have the date/time when they opened the mail . ...
1
vote
2answers
22 views

Multiple models have extreme differences during evaluation

My dataset has about 100k entries, 6 features, and the label is simple binary classification (about 65% zeros, 35% ones). When I train my dataset on different models: random forest, decision tree, ...
1
vote
1answer
24 views

How can I model this problem to train a Machine learning model

I am new to Machine Learning and need help understanding how can I model the below (hypothetical) problem. Say I record the following info about thousands of people. For 365 days, each day I record ...
1
vote
1answer
11 views

Tune learning rate while tuning other HP

When doing hyperparameters optimisation, like a Random Search, should you add a search space for the learning rate ? My intuition is that some HP might work better with a certain LR, and be sub-...
0
votes
1answer
33 views

when I only give command 'fit', my class does 'transform' too

I have created 2 classes, first of which is: ...
0
votes
0answers
7 views

VC Dimension of some certain Hypothesis Classes

I'm still unsure as to how to get the VC-dimension of some certain hypothesis classes. So for example we have $\mathcal{X} = \mathbb{R}$ and $\mathcal{Y} = \{ 0,1 \}$ and the hypothesis class $\...
0
votes
2answers
37 views

Supervised clustering use case?

I'm currently working in a problem, where I think a supervised clustering approach might be a good candidate, but I'm not sure and haven't really worked with such scenario before. Let me break it down:...
1
vote
1answer
32 views

What approach should I take for my product classification ML model with user feedback for improving result accuracy?

I'm trying to implement a product categorization ML model on a dataset with the following structure: Data sample I want to my model to be able to predict the correct category that the product should ...
1
vote
0answers
10 views

How to create transition probability (state) for q-learning algorithm designed to control traffic light system using python?

I am trying to create a q learning algorithm to control traffic light systems. I am representing the state with a matrix : ...
0
votes
0answers
44 views

My Models giving negative scores

I am new to Data Science. I am trying to use following dataset in order to predict prices for some reason my models except for decision tree is giving negative score. Please help me to build this ...
0
votes
0answers
10 views

How to represent source code in NLP for tasks like code retrieval?

How to represent code features in NLP? Can code be treated like a language and pour into the neural network? Are there some work, paper or material around this topic?
0
votes
0answers
20 views

Understanding how transfer learning happens in named entity recognition task

I was going through word embedding video in Andrew Ng's coursera course Sequence modeling. In this video, he gives following two examples: Sally Johnson is an orange farmer. Robert Lin is a durian ...
0
votes
2answers
52 views

Removing illogical observations, can I do it?

So I have data that has some illogical entries, in the picture below you can see a person with 11 years in the company and he takes 100 dollars, this is for sure wrong and should be removed right? The ...
1
vote
1answer
38 views

Why does stochastic gradient descent lead us to a minimum at all?

Why do we think that stochastic gradient descent is going to find a minimum at all? I mean on each iteration SGD moves in the direction that reduces only current batch's error (SGD doesn't care about ...
1
vote
3answers
47 views

Does One-Hot encoding increase the dimensionality and sparsity of dataset?

There are two ways to convert object datatype into numeric datatype, first is One-Hot encoding and second is simply map the numerical tags to different values. For example for column Age containing ...
1
vote
1answer
22 views

Normalization factor in logistic regression cross entropy

Given that probability of a matrix of features $X$ along with weights $w$ is computed: ...
1
vote
1answer
15 views

Understanding deconvolutional network loss function

In the paper (1), there is a description of a deconvolutional network. The loss function (with only one layer) compares the colour channels of the orignal image with the colour channels of the ...
0
votes
0answers
17 views

Is it always appropriate to use SMOTE in an imbalanced multiclass dataset?

Is it good practice to always use SMOTE and random undersampling in an imbalanced multiclass dataset or are there exceptions? In context, I am using a traditional machine learning model (SVC) for ...
2
votes
2answers
104 views

Removing the outliers improved my models, is what I did good or bad?

I used cross validation on my data (11000 rows) with maximum salary of 10000 and after some cleaning I got to rmse=70. Then I tried to remove the outliers 10 times just to try things now I have 9000 ...
0
votes
0answers
12 views

The explicit event which gives distinguisher between two distributions

I am very very new in machine learning. Suppose two distributions generate binary numbers. For the 1st distribution, 0 occurs with a probability of 0.5; for 2nd, it is 0.51. I generate data from both ...
0
votes
1answer
11 views

How many layers of a pretrained model shoud be frozen?

I'm following an example of transfer learning where the blogger has frozen the first 20 layers of MobileNet. My question is , that is there any rule of thumb for how many layers should be frozen? ...
0
votes
1answer
13 views

Evaluate Text-to-speech without Human Involved?

I've explored text-to-speech evaluation matrices and they seem to used Mean Opinion Score (MOS) to evaluate a particular model. This matrice required humans to help to judge the model based on a scale ...
1
vote
0answers
6 views

Legality/common practice when calling APIs in an R package?

Without getting into too much detail, for a personal project, I have built a model in R that will call Google Maps Platform API see here to pull driving times for a user’s given input and perform a ...
7
votes
1answer
94 views

Why deep learning models still use RELU instead of SELU, as their activation function?

I am a trying to understand the SELU activation function and I was wondering why deep learning practitioners keep using RELU, with all its issues, instead of SELU, which enables a neural network to ...
0
votes
0answers
33 views

How to split Algorithms into multiple machines? [closed]

I have NLP big dataset (millions of records) and I am trying to find best algorithm that can do predictive modeling / classifications I use these algorithms mainly (Logistic Regression, Random Forest, ...
0
votes
1answer
26 views

How do I calculate precision, recall, specificity, sensitivity manually?

I have actual class labels and predicted class labels: ...

1
2 3 4 5
194