95 interview questions for developers (2025)

Medium

18 votes1370 runs0 comment

How to handle missing data in a given training data set ?

Data Scientist

machine learning

missing data

Martinat Apr 4, 2022

Medium

11 votes1331 runs0 comment

The validation set is used to provide frequent and unbiased evaluations of the model’s fit on the training set while tuning its hyperparameters/parameters: in other words,the model is found and then tested on the validation set before to be improved once again.

Data Scientist

hyperparameters

model

Martinat Mar 15, 2022

Medium

10 votes1237 runs0 comment

How to handle missing data in a given training data set ?

Data Scientist

machine learning

missing data

Martinat Apr 4, 2022

Easy

10 votes896 runs0 comment

How is the error calculated during back propagation in a neural network?

Data Scientist

neural network

back propagation

Anonymous authorat Apr 6, 2022

Easy

7 votes1259 runs0 comment

What is the missing word in the following sentence: Overfitting is the production of a model that corresponds too closely to the training data set and may therefore fail to fit other data sets and so fail to perform any reliable forecast.

Data Scientist

machine learning

overfitting

Martinat Mar 15, 2022

Hard

5 votes1210 runs

2 comments

What are the differences/similarities between a loss function, an error function, and a cost function?

Data Scientist

loss function

error function

cost function

Martinat Mar 15, 2022

Easy

5 votes1140 runs0 comment

Clustering is an unsupervised machine learning process which aims at automatically discovering natural grouping in the input data. When using this method, the developer is able to choose the number of groups he wants to create.

Data Scientist

machine learning

unsupervised learning

Martinat Mar 15, 2022

Easy

5 votes833 runs0 comment

What is ensemble learning? It consists in …

Data Scientist

models

ensemble

Martinat Mar 15, 2022

Easy

4 votes1273 runs0 comment

Given a data set, which elements will help you to choose what model / machine learning method to use to create an efficient model?

Data Scientist

data

model

Martinat Mar 15, 2022

Easy

4 votes1191 runs0 comment

In order to get good results when implementing deep learning, one needs to prepare his data before using it. What action should be done during this preparation?

Data Scientist

data

preprocessing

Martinat Mar 15, 2022

Medium

4 votes1157 runs0 comment

How to assess a clustering model performances?

Data Scientist

metrics

clustering

Martinat Mar 15, 2022

Easy

4 votes844 runs0 comment

In the context of NLP, what is the role of the Stemming method?

Data Scientist

nlp

stemming

Anonymous authorat Apr 6, 2022

Easy

4 votes777 runs1 comment

What type of data is PCA (principal component analysis) applied to?

Data Scientist

data

machine learning

Anonymous authorat Apr 6, 2022

Hard

4 votes739 runs0 comment

How are the layers of an MLP type neural network connected to each other?

Data Scientist

neural networks

linear

Anonymous authorat Apr 6, 2022

Medium

3 votes1192 runs0 comment

Which of the following statements is true when considering machine learning?

Data Scientist

machine learning

artificial intelligence

computer science

Martinat Mar 15, 2022

Easy

3 votes1138 runs0 comment

How to define supervised learning? In supervised learning, the machine trains the model using a labeled training data set.

Data Scientist

supervised learning

machine learning

Martinat Mar 15, 2022

Easy

3 votes785 runs0 comment

In an MLP(multilayer perceptron), the output layer necessarily has a single neuron.

Data Scientist

neural networks

multilayer perceptron

Anonymous authorat Apr 6, 2022

Easy

3 votes759 runs0 comment

SVMs seek to find a hyperplane that maximizes the width of the separating band between data of different classes.

Data Scientist

machine learning

svm

Anonymous authorat Apr 6, 2022

Hard

2 votes1195 runs0 comment

How to define reinforcement learning? Select the sentence that defines best this learning method.

Data Scientist

reinforcement learning

machine learning

Martinat Apr 4, 2022

Easy

2 votes805 runs1 comment

What does the Soft Margin version of SVMs do?

Data Scientist

machine learning

svm

Anonymous authorat Apr 6, 2022

95 interview questionsfor developers (2025)

Test your developer skills before an interview with 95 corrected questions on 80+ topics. Beginner, Intermediate or Expert level.

How to handle missing data in a given training data set ?

The validation set is used to provide frequent and unbiased evaluations of the model’s fit on the training set while tuning its hyperparameters/parameters: in other words,the model is found and then tested on the validation set before to be improved once again.

How to handle missing data in a given training data set ?

How is the error calculated during back propagation in a neural network?

What is the missing word in the following sentence: Overfitting is the production of a model that corresponds too closely to the training data set and may therefore fail to fit other data sets and so fail to perform any reliable forecast.

What are the differences/similarities between a loss function, an error function, and a cost function?

Clustering is an unsupervised machine learning process which aims at automatically discovering natural grouping in the input data. When using this method, the developer is able to choose the number of groups he wants to create.

What is ensemble learning? It consists in …

Given a data set, which elements will help you to choose what model / machine learning method to use to create an efficient model?

In order to get good results when implementing deep learning, one needs to prepare his data before using it. What action should be done during this preparation?

How to assess a clustering model performances?

In the context of NLP, what is the role of the Stemming method?

What type of data is PCA (principal component analysis) applied to?

How are the layers of an MLP type neural network connected to each other?

Which of the following statements is true when considering machine learning?

How to define supervised learning? In supervised learning, the machine trains the model using a labeled training data set.

In an MLP(multilayer perceptron), the output layer necessarily has a single neuron.

SVMs seek to find a hyperplane that maximizes the width of the separating band between data of different classes.

How to define reinforcement learning? Select the sentence that defines best this learning method.

What does the Soft Margin version of SVMs do?