DSE Basic 3 — Questions and Answers
Question 1: What is a 'feature' in the context of machine learning?
- The target variable to predict
- An individual measurable property used as model input (Correct answer)
- A model's accuracy score
- A hyperparameter setting
Correct answer: An individual measurable property used as model input
A feature is an individual measurable input variable that the model uses to make predictions.
Question 2: Which type of chart is best for showing the distribution of a single continuous variable?
- Pie chart
- Bar chart
- Histogram (Correct answer)
- Scatter plot
Correct answer: Histogram
A histogram groups continuous data into bins and shows frequency, revealing the distribution shape.
Question 3: What does SQL stand for?
- Structured Query Language (Correct answer)
- Sequential Query Logic
- Standard Query Layer
- Stored Query List
Correct answer: Structured Query Language
SQL stands for Structured Query Language, used to manage and query relational databases.
Question 4: In supervised learning, what is the 'label'?
- A feature column name
- The known output used during training (Correct answer)
- The model's predicted value
- A data cleaning step
Correct answer: The known output used during training
The label (or target) is the known correct output that the model learns to predict during training.
Question 5: What is the interquartile range (IQR)?
- Q1 minus the minimum value
- Q3 minus Q1 (Correct answer)
- The mean minus the median
- Q4 minus Q2
Correct answer: Q3 minus Q1
The IQR is Q3 − Q1, representing the middle 50% spread of data and useful for detecting outliers.
Question 6: Which of the following is an example of categorical data?
- Temperature in Celsius
- Number of sales per day
- Customer satisfaction level (Low/Medium/High) (Correct answer)
- Annual revenue in dollars
Correct answer: Customer satisfaction level (Low/Medium/High)
Categorical data represents discrete groups or categories rather than numerical measurements.
Question 7: What is overfitting in machine learning?
- The model performs poorly on both training and test data
- The model learns training data too well and fails to generalize (Correct answer)
- The model is too simple to capture patterns
- The model trains too slowly
Correct answer: The model learns training data too well and fails to generalize
Overfitting occurs when a model memorizes training data noise, causing poor performance on unseen data.
DSE Basic 3
What is a 'feature' in the context of machine learning?