Data Science Professional Certification

1. Which of the following algorithms is typically used for classification tasks in data science?

K-Means Clustering

Linear Regression

Support Vector Machine (SVM)

Apriori Algorithm

2. Which of the following is a supervised learning algorithm?

Principal Component Analysis (PCA)

Decision Tree

K-Means

DBSCAN

3. In Natural Language Processing (NLP), which technique is used to convert text data into numerical vectors?

Bag of Words (BoW)

Random Forest

Gradient Descent

Boosting

4. Which metric is NOT typically used to evaluate a classification model?

Precision

Recall

F1 Score

Mean Squared Error (MSE)

5. Which type of cross-validation involves dividing the data into "k" subsets and using each subset as a test set once?

Holdout Validation

K-Fold Cross-Validation

Leave-One-Out Cross-Validation (LOOCV)

Random Sampling

6. Which Python library is commonly used for data manipulation and analysis?

Matplotlib

NumPy

Pandas

Scikit-learn

7. Which of the following is an example of an ensemble learning method?

Linear Regression

K-Nearest Neighbors (KNN)

Random Forest

Naive Bayes

8. What does the “curse of dimensionality” refer to in data science?

Difficulty in interpreting too much data

The computational cost of processing high-dimensional data

Overfitting due to too many features

Loss of data during feature extraction

9. Which activation function is commonly used in hidden layers of deep neural networks?

Sigmoid

ReLU (Rectified Linear Unit)

Softmax

Linear

10. What is the correct way to handle events in React?

Using inline event handlers with strings.

Directly modifying the DOM with JavaScript.

Using event handlers as functions passed to JSX elements.

Using the on prefix for attribute names, like onClickEvent.

10. In a dataset with a large number of missing values, what is the first step a data scientist should consider?

Delete all rows with missing values

Impute missing values with mean or median

Analyze the reason for missing data

Replace missing values with zeros

12. Which machine learning technique is most appropriate for predicting a continuous numeric target variable?

Logistic Regression

Linear Regression

K-Means Clustering

Support Vector Classification (SVC)

13. Which of the following techniques is used to reduce the dimensionality of a dataset?

Linear Regression

Principal Component Analysis (PCA)

K-Means Clustering

Logistic Regression

14. In the context of machine learning, what does "underfitting" mean?

The model performs poorly on both training and test data

The model performs well on training data but poorly on test data

The model performs equally well on training and test data

The model performs extremely well on test data but poorly on training data

15. What is the purpose of the "learning rate" in gradient descent?

To control the number of iterations in the algorithm

To control the speed at which the algorithm converges to the minimum

To scale the data before feeding it into the model

To determine the size of the input data

16. Which of the following is a type of unsupervised learning?

Support Vector Machine (SVM)

K-Nearest Neighbors (KNN)

K-Means Clustering

Logistic Regression

17. Which of the following is true about the "Bias-Variance Tradeoff" in machine learning?

High bias leads to a model that is too complex

High variance leads to a model that generalizes well to new data

A model with both high bias and high variance will underperform

Reducing bias always improves model performance

18. Which metric is commonly used to evaluate the performance of a regression model?

Confusion Matrix

Area Under the Curve (AUC)

Mean Squared Error (MSE)

Precision

19. What is the key idea behind the "Naive Bayes" algorithm?

It assumes all features are dependent on each other

It uses a decision tree to predict outcomes

It applies Bayes' Theorem with an assumption of independence between features

It iteratively refines the classification boundaries using gradient descent