1️⃣ Question 1

Which describes a file with plain text, rows, and columns?

• ❌ Key–value text

• ❌ Comma-separated array

• ❌ Excel spreadsheet

• ✅ A text file that saves data in tables

Explanation:
A plain text table (e.g., .txt or .csv) is stored as rows/columns.

---

2️⃣ Question 2

Python library for email spam classification?

• ❌ Fast array processing

• ❌ Exploratory data analysis

• ❌ Matrix operations

• ✅ Statistical modeling including regression/classification (scikit-learn)

Explanation:
Scikit-learn provides classification algorithms.

---

3️⃣ Question 3

Most important factors for reading data with Pandas:

• ❌ Encoding + file path

• ❌ File types + formats

• ❌ Format + file path

• ✅ File types and encoding scheme

Explanation:
Pandas must know file type (CSV/JSON/Excel) and encoding (UTF-8, ISO-8859-1).

---

4️⃣ Question 4

Why use Python’s DB API?

• ❌ Autogenerate interfaces

• ❌ Bypass SQL

• ❌ Convert output to JSON

• ✅ Allows consistent querying/connection across SQL systems

Explanation:
Python DB API standardizes database connections.

---

5️⃣ Question 5

Regression model of: ŷ = b₀ + b₁x

• ✅ Linear regression

• ❌ Polynomial

• ❌ Multiple linear

• ❌ Exponential

---

6️⃣ Question 6

Why do predictions become negative?

• ❌ Coefficients uninterpreted

• ✅ Model extrapolates beyond realistic ranges

• ❌ Regression line always increases

• ❌ Low R²

Explanation:
Linear models fail outside the data range.

---

7️⃣ Question 7

Challenge of a single train–test split:

• ❌ Invalid R²

• ❌ Less accuracy due to less training data

• ✅ Generalization error may change with each split

• ❌ Decline to adapt to hidden features

Explanation:
One split may not represent the dataset; different splits yield different performance.

---

8️⃣ Question 8

Code effect:

• ❌ Only calculates mean

• ✅ Fills missing values with column mean

• ❌ Normalizes data

• ❌ Drops missing rows

---

9️⃣ Question 9

Purpose of binning car prices:

• ❌ Randomizes prices

• ❌ Filters values

• ✅ Creates labeled segments for price intervals

• ❌ Equalizes values

---

🔟 Question 10

Method dividing by standard deviation:

• ❌ Simple scaling

• ✅ Z-score standardization

• ❌ Feature binning

• ❌ Min-max scaling

---

1️⃣1️⃣ Question 11

Method to evaluate column statistics/types:

• ❌ dataframe.values()

• ❌ dataframe.rename()

• ❌ dataframe.astype(“int”)

• ❌ dataframe.dtypes(“int”)
  Correct: Pandas method is dataframe.dtypes (last option is incorrectly written but intended).
  So the correct choice is:

• ✅ dataframe.dtypes(“int”) (intention: check data types)

---

1️⃣2️⃣ Question 12

What is EDA?

• ✅ Reviewing key characteristics and uncovering patterns

• ❌ Segmenting dataset

• ❌ Minimizing dimensions

• ❌ Training models

---

1️⃣3️⃣ Question 13

Negative linear relationship means:

• ❌ Output doesn’t explain input

• ❌ Decreases at increasing rate

• ✅ With increase in input, output decreases at about the same rate

• ❌ Output increases

---

1️⃣4️⃣ Question 14

Method to find outliers:

• ❌ Scatter plot

• ✅ Box plot

• ❌ Histogram via describe

• ❌ value_counts

---

1️⃣5️⃣ Question 15

Method to compare average price across drive types:

• ✅ Group the data using category values

• ❌ Numeric filters

• ❌ Row filters

• ❌ Combine datasets

Explanation:
Use groupby().

---

1️⃣6️⃣ Question 16

Role of independent variables:

• ❌ Summarize performance

• ❌ Define accuracy metric

• ❌ Compare models

• ✅ Serve as inputs to estimate output

---

1️⃣7️⃣ Question 17

Curved residual pattern implies:

• ❌ Linear relationship

• ❌ Uniformly low errors

• ❌ Randomly distributed

• ✅ Model may be inaccurate → nonlinear relationship

---

1️⃣8️⃣ Question 18

Truth about noise:

• ❌ Accounted with a parameter

• ❌ No noise if testing fits

• ❌ No noise if training fits

• ✅ Noise is random and cannot be predicted

---

1️⃣9️⃣ Question 19

Large alpha in ridge regression:

• ❌ Lower-order function required

• ✅ Model is underfitted

• ❌ Overfitted

• ❌ Higher alpha → better fit

Explanation:
High alpha shrinks coefficients too much → underfitting.

---

2️⃣0️⃣ Question 20

Argument passed to GridSearchCV():

• ❌ Dictionary of columns

• ✅ Dictionary of parameters and values

• ❌ Dataframe of models

• ❌ Normalized features

---

🧾 Summary Table

---