In one-factor ANOVA, the residuals:

Are assumed to be correlated

Are assumed to be normal

We do not need any assumptions on the residuals

The null hypothesis in a one-factor ANOVA states that:

All the means are different

All the means are the same across levels

There are only two means that are the same

The two-factor ANOVA statistic to test for the significance of factor 𝜶 has a distribution:

Snedecor’s F

Chi-square

Student’s t

Which of the following most closely describes the machine learning approach to completing a task?

Keep trying at random until successful

Follow step-by-step instructions

Detect a pattern from historical data or previous trials and apply it

Which algorithm assumes that input features are mutually independent?

Neural Networks

Naïve Bayes

Decision Trees

How many partitions is your training dataset divided into in a 5-fold CV? And how many times does the algorithm get trained in total?

4 partitions; 5 trainings

5 partitions; 6 trainings

5 partitions; 5 trainings

Name three different examples of using data science for good

monitoring social media for human rights impacts

adaptive charging

skills matching

Which of the following is not one of the principles of trustworthy AI?

Reproducibility

Transparency

Robustness

The Equalized Odds fairness metric requires that

All of the above

FPR is the same for all demographic groups

TPR is the same for all demographic groups

Typical text mining tasks include:

Text categorization

Modeling entity relation

All of the above

Text clustering

Is stemming the process of separating the prefixes and suffixes from words to derive the root word form and meaning?

True and False

True

False

Which one of the following word embeddings can be custom trained for a specific subject in NLP?

Word2Vec

GloVe

BERT

All the above