Questions tagged [feature-engineering]
the process of using domain knowledge of the data to create features that improve machine learning algorithms
443
questions
1
vote
1answer
18 views
Machine Learning with intended missing values
I have a dataset relating to humans completing reviews, the target variable is whether the review decision is correct / incorrect and one of my features is a trailing 4 week accuracy score for the ...
0
votes
0answers
14 views
How to select variables in panel data? [closed]
So in setting of non-panel data (i.e. one row per individual) I would check the correlations between variables and discard some which cause high correlation. How shall I check and select when the data ...
2
votes
1answer
17 views
Cannot understand feature extraction
I'm following an AI course and we've just entered the deep learning chapter. Speaking about the difference between classic machine learning models and deep learning, it turns out one of the most ...
1
vote
0answers
14 views
Feature engineering and longitudinal data
I need some advice for my feature engineering.
Suppose I have 90 days follow-up data. on 12 patients and I have the vital status of the patients at the end of these 90 days (deceased=1, alive=0)
...
1
vote
1answer
16 views
What is a good approach for embedding both textual and spatial features for document classification?
I am working on a document classifier that can perform the classification based on the document structure as well. My plan is to get the word embedding as well as the word coordinates and somehow ...
0
votes
0answers
4 views
What are good feature selection and engineering approaches for data with known uncertainties?
Context: I am working with a set of geological features that could have uncertainty values attached to them (for example, values come from drill holes that are sparsely distributed and must be ...
0
votes
0answers
13 views
Feature Engineering and prediction with R and python
I have a sequential dataset, and have 2000 rows for 300 ID. I have 20 variables (i.e in my real dataset)
...
1
vote
3answers
36 views
Is it possible to change the input columns of a trained ML model while making predictions from it without affecting the accuracy?
Consider the following scenario.
I have trained a K-Means model on some input features, say, (A, B, C, D and E).
Now at the time of making predictions I want to make the model predict using only fewer ...
1
vote
1answer
39 views
How to model a supervised recommender system with varying data
Suppose there are 2000 movies and a company wants to recommend some movies (for example, at most 5 movies) to each visitor. The objective is to learn how to predict which movie will be selected if a ...
2
votes
0answers
31 views
Should you use the same algorithm in your feature selection as your model
My question is should you use the same algorithm in feature selection as your model?
If I'm using a KNN model for classification should I also use a KNN algo when running feature selection? Or ...
0
votes
0answers
18 views
how do i preprocess percentage data?
I am analyzing a problem where i have 5 diseases and the measure of effectiveness of a remedy in its 7 first applications. The data is organized as an Excel spreadsheet as follows:
(The spreadsheet ...
3
votes
1answer
32 views
How best to use the resale transaction year in predicting housing prices?
I'm looking into the classic problem of predicting apartment prices (resale market) based on the their type, size, location, etc. Pretty straightforward and Linear Regression or Regression Trees give ...
0
votes
0answers
13 views
Converting the continuous numerical features into gaussian distribution and how to deal with NaN values after that?
I have a dataset in which there are few continuous numerical features that distribution over them is non gaussian and this means, skewness is nonzero (positive or negative).
I read that it is good to ...
0
votes
0answers
18 views
Can we custom / teach machine learning to put-attention at a feature or make a feature the most importance in model?
I have real dataset that contains place, address, and category of two places. I will match those name and address but I will give exception based the category column.
This the example of dataset ...
0
votes
1answer
23 views
Methods of disaggregating data to smaller units?
I have a relatively straightforward question that I know poses some difficult challenges.
Let's say I have a state-level rate of X. I would like to disaggregate the state-level rate to the county-...
1
vote
0answers
32 views
Is there a standard method for choosing features from different feature selection techniques?
I have four different feature selection techniques, Backwards Elimination, Lasso, feature_importances, and Recursive feature selection. Each technique returns slightly different results. For example,
...
3
votes
1answer
37 views
How data are prepared during training, testing and in production?
Most of real world datasets have features with missing values. Replacing missing values with an appropriate value such as its mean, is considered as a good step in feature engineering. Some times we ...
0
votes
0answers
7 views
MSE errors on autoencoder for dim reduction decreases in a weird patteren and I would love some help to dechyper it
I'm training a denoising autoencoder right now to reduce the dimension of a feature vector I designed of dim 58 to a latent space of dim 10, or less hopefully. I'm having a hard time understanding ...
0
votes
0answers
3 views
Does TargetEncoder class from category_encoders include cross validation
I learnt (from below links) that to effectively do target encoding without overfitting, we have to do cross validation for each fold (so kind of double validation) and compute the encoding values of ...
1
vote
0answers
34 views
Examples of “unusual”/non-trivial features that actually worked for improving model score [closed]
I have been working for a while in credit problems for classification and regression and on these problems I have had the necessity of improving already good performing models, for this when ...
1
vote
1answer
19 views
Modern methods for reducing dimensions and feature engineering
I am training a binary classifier in Python to estimate the level of risk of credit applicants. I extracted a little over a thousand independent variables to model the observed behavior of four ...
0
votes
0answers
14 views
What are the different ways to feature engineer webpage data for input into a webpage classification model?
Looking for resources on the different ways that one can manipulate webpage data to input as features into a neural net.
I'm aware of a service called diffbot that claims to use a CV based method to &...
2
votes
1answer
43 views
How modelling is affected by similar feature distributions across classes?
I have grouped my dataset according to labels (good and bad customers), in an attempt to test how each feature is distributed within.
Along this, I found some feature has almost exact distribution in ...
0
votes
0answers
59 views
When to consider examples as duplicates in a classification problem?
I'm trying to train a model for a binary classification problem and I'm trying to understand if my model is biased from my data or if my results are valid.
I have a pandas dataframe as output of my ...
0
votes
1answer
13 views
sine, cosine transformed cyclical features - am I losing information?
If I use sine, cosine transformation for cyclical features (e.g. weekday or hour of the day), do I lose information if the first ordinal value was 0 respectively?
Assume hours of the day are encoded ...
0
votes
0answers
26 views
Is feature weighting is helpful for boosting Machine Learning model performance?
I am a beginner in machine learning. I am looking to know, ...
0
votes
0answers
7 views
Create a single feature vector from the 2 edges of the vertex
Overview
Consider the 4 examples of the right angled vertex shown below. In each example, the vertex is made up of 2 line segments- A and B, which are perpendicular, and the all examples are nearly ...
0
votes
0answers
15 views
Input variable that influences only trend
I need to predict cost for the next 4 weeks. Along with categorical variables (available for future as well), I have a numerical variable (a value from the day before prediction - I don't have future ...
0
votes
1answer
38 views
How to input sets as features
Need advice on the best way to represent the below data to be fed into an ML algorithm (yet to decided on)
This is from the online order processing domain. An order consists of a set of variable ...
0
votes
0answers
11 views
Usage of Doc2Vec as feature extractor for text classification of websites with political articles
I have gathered political articles from polish websites for my engineering thesis. The main goal is to try to predict the website that input text belongs to. So for this few websites I want to create ...
0
votes
1answer
19 views
Generate new features from two columns
I have database with three columns, y,x1 and x2:
>>>y x1 x2
0 0.25 -19.3 -25.1
1 0.24 -18.2 -26.7
2 0.81 -45.2 -31.4
...
I want ...
1
vote
1answer
19 views
How to deal with variable number of permutation invariant features?
I want to learn from data where each record has a variable number of features that have no inherent order to them. Take as an example the task to predict whether a repair is worth it of some item. ...
0
votes
0answers
12 views
Determine most important features per class in binary classification?
The only good answer i found was this :
https://stackoverflow.com/questions/33118361/determine-most-important-feature-per-class
But the above answer doesn't work in binary classification because ...
0
votes
2answers
16 views
Feature reduction by removing certain columns in dataframe
I am working with the Emotion recognition model with the IEMOCAP dataset. For the feature extraction, I am taking mel-spectrogram and then convert it into a NumPy array and converting the array into a ...
0
votes
1answer
13 views
What to do this type of data
I have plots of feature vs target values as
and
And many more. I have used Tree based models and able to get high accuracy but not able to improve further. Any suggestion ? almost every feature is ...
0
votes
0answers
20 views
How should I encode 'dynamic' features (with multiple instances) along with 'static' features (single instances)?
Suppose I have to predict if a certain product from an assembly line in a factory will be a scrap. This product has let's say 'static' data like a certain shape. A certain vendor, etc. And, it can ...
0
votes
1answer
18 views
Is the result of feature extraction a feature representation?
If a use a feature extraction method on images, do I then get a feature representation or is there a different meaning behind feature representation?
To my understanding, when I use a CNN on an image ...
0
votes
0answers
21 views
Noisy features detection
Based on the library featexp I am trying to identify noisy categorical features. I want to know if this is the right way and if there is some library for this solution it will be great to know it.
I ...
2
votes
1answer
45 views
Can we use two independently measured features in a same ML model? [closed]
Two features are measured at different times but belong to the same target. In which conditions or form these features can be modeled together? Or they shouldn't be used in the same model but modeled ...
0
votes
1answer
11 views
What is best practice to feature engineer from prior event counts?
Say for example I am building a model to predict a customer churn event from Spotify, with my target being whether a customer churns in the next 90 days.
One feature I might expect could be predictive ...
2
votes
0answers
15 views
Encoding ML classification features that are relative to the dependant categories
I have a classification problem with three classes A, B & C.
I have features x1, x2 and x3 that represent my data items. But I also have a fourth feature that represents a similarity between my ...
0
votes
2answers
33 views
Problem with a feature (normal distribution + peak around 0)
I have a feature that shows a characteristic of the instances. That characteristic can be present or not. If present it shows an almost normal distribution of values (actually a bit skewed to the ...
3
votes
2answers
43 views
Dealing with highly variable feature set size
I'm trying to use machine learning for security event classification. My goal is to predict the outcome (true positive or false positive) of a specific event. An event has a set of variables in it, ...
-1
votes
1answer
31 views
Is there any advantage of limiting the value of a feature in neural networks
In a machine learning algorithm, I have a feature that has a value in the range 0-20 it is very rarely value goes over 20 and if does I clamp it 20.
Does it help the neural network model somehow using ...
0
votes
1answer
21 views
Reduce Categorical Values
I'm working on one use case where I have to explore source code repo files. Different files will be a categorical values for me. But with such large number of files, One Hot Encoding comes out to be ...
1
vote
0answers
31 views
Handling highly correlated features [closed]
I have a data set of transactions and want to build a fraud detection model (classifier). Only 3 variables are given that could be used as input features. The number of transactions during past 3, 6 ...
1
vote
1answer
30 views
How to handle a valuable feature that is missing on 99\% of the samples in the data set?
Suppose we have an input feature that is highly predictive of the outcome we want to predict. However, the feature is missing on 99% of the samples in the data set. What is the best way to use this ...
1
vote
1answer
21 views
Drastic shift in feature importance upon adding other features
I have a model (GBDT) where adding a feature X is not important (according to SHAP), but when I add other features, and add X again, now feature X is the second most important!
What could explain that?...
1
vote
2answers
38 views
Can machine learning models treat a vector as a whole feature to learn
We know a ML model naturally takes a feature vector with real valued elements as input and learn to predict. But can it treat a fixed-size vector as a whole feature to learn?
For example, when using a ...
0
votes
1answer
38 views
Linear and non-linear dependence in a single DS model
I have a dataset with parameters (features) a,b,c, etc. We need to develop a model to ...
