Here are the correct answers with brief explanations:
1. c) 0
[Link]('c', 0) returns the default value 0 since key 'c' is not in the dictionary.
2. b) if
The if keyword is used for conditional branching in Python.
3. a) array()
NumPy uses array() to create arrays.
4. c) Dimensions of array
[Link] returns the dimensions (rows, columns, etc.) of the array.
5. d) drop_duplicates()
In Pandas, drop_duplicates() removes duplicate rows.
6. b) 1D labeled array
A Pandas Series is a one-dimensional labeled array.
7. c) Pickle
Pickle is a binary format used for efficient storage and serialization.
8. b) HTTP
Web APIs typically use the HTTP protocol.
9. d) Pivoting and reshaping
These techniques restructure datasets for reporting and analysis.
10. c) Complex relationships across multiple dimensions
Multi-level indexing helps analyze data across multiple dimensions.
If you want, I can turn these into a printable quiz or add more MCQs for practice.
11) State the purpose of built-in sequence functions. (RE) 12) Define namespace and scope. (RE)
13) Differentiate between Python list and NumPy array. (RE) 14) How would you generate a
sequence of pseudorandom numbers?(RE) 15) Explain about the descriptive statistics? (UN) 16)
Justify the best method for handling missing values. (RE) 17) Short Notes: Merging of datasets?
(RE) 18) How pivot tables summarize data. (UN) 19) Assess the effectiveness of EDA methods
(EDA)? (RE) 20) What is scatter plot? (RE)
Here are concise answers based on the required taxonomy levels:
11) Purpose of built-in sequence functions (RE)
Built-in sequence functions (like len(), sum(), min(), max(), sorted()) are used to perform
common operations on sequences (lists, tuples, strings) such as counting elements, finding
extremes, sorting, and aggregating values.
12) Namespace and scope (RE)
A namespace is a container that holds names (identifiers) and their corresponding objects.
Scope defines the region of a program where a namespace is accessible (local, global, enclosing,
built-in).
13) Python list vs NumPy array (RE)
List: Can store different data types, slower, flexible.
NumPy array: Stores same data type, faster, supports vectorized operations, efficient for
numerical computations.
14) Generating pseudorandom numbers (RE)
Use the random module or NumPy:
[Link]() (Python)
[Link]() or [Link]() (NumPy)
15) Descriptive statistics (UN)
Descriptive statistics summarize and describe data features using measures like mean, median,
mode, standard deviation, variance, minimum, and maximum. They help understand data
distribution and patterns.
16) Handling missing values (RE)
Best methods include:
Removing missing values (if small amount)
Filling with mean/median/mode (imputation)
Forward/backward filling
Choice depends on data size and importance of missing values.
17) Merging of datasets (RE)
Merging combines multiple datasets based on common keys or columns using operations like
inner join, outer join, left join, and right join (e.g., [Link]()).
18) Pivot tables (UN)
Pivot tables summarize data by reorganizing it, grouping values, and applying aggregate
functions (sum, mean, count) to generate insights across categories.
19) Effectiveness of EDA (RE)
EDA (Exploratory Data Analysis) is effective for:
Understanding data structure
Detecting patterns and trends
Identifying outliers and anomalies
Supporting better decision-making before modeling
20) Scatter plot (RE)
A scatter plot is a graph that displays values for two variables as points, used to identify
relationships or correlations between them.
21(a) Evaluate performance of arrays, lists, and dictionaries (EV)
Lists (Python):
Time Complexity:
o Access: O(1)
o Insert/Delete (end): O(1), (middle): O(n)
Space: Dynamic, can store mixed data types → less memory efficient
Use case: General-purpose, flexible storage
NumPy Arrays:
Time Complexity: Faster for numerical operations due to vectorization (O(n) but
optimized in C)
Space: Memory efficient (homogeneous data)
Use case: Scientific computing, large datasets
Dictionaries:
Time Complexity:
o Access, Insert, Delete: O(1) (average, hash-based)
Space: Higher memory usage due to hash table
Use case: Fast lookup using keys
Evaluation:
Arrays → best for performance & memory in numerical tasks
Lists → flexible but slower
Dictionaries → fastest for key-based retrieval
21(b) Functions improve modularity with file handling (AN)
Functions divide programs into reusable blocks, improving readability and maintenance.
Benefits:
Code reuse
Easier debugging
Better organization
Example:
def read_file(filename):
with open(filename, 'r') as f:
return [Link]()
def write_file(filename, data):
with open(filename, 'w') as f:
[Link](data)
data = read_file("[Link]")
write_file("[Link]", [Link]())
Analysis:
Each function performs a specific task
Changes in one function don’t affect others
Enhances modular design
22(a) Universal functions in NumPy (UN)
Universal functions (ufuncs) operate element-wise on arrays.
Features:
Fast (implemented in C)
Support broadcasting
Work on entire arrays
Examples:
import numpy as np
arr = [Link]([1, 2, 3])
print([Link](arr)) # [1. , 1.41, 1.73]
print([Link](arr, 5)) # [6, 7, 8]
Explanation:
No need for loops
Improves performance and readability
22(b) Linear algebra functions in NumPy (EV)
Common functions:
[Link]() → matrix multiplication
[Link]() → inverse
[Link]() → determinant
[Link]() → eigenvalues
Evaluation:
Efficient and accurate for matrix operations
Handles large-scale problems
Limitation: Numerical instability in some cases
Example:
import numpy as np
A = [Link]([[1,2],[3,4]])
print([Link](A))
print([Link](A))
23(a) Python app: Web API → Database (CR)
Steps:
1. Fetch data from API
2. Process JSON
3. Store in database
Example:
import requests
import sqlite3
# Fetch API data
response = [Link]("[Link]
data = [Link]()
# Store in database
conn = [Link]("[Link]")
cursor = [Link]()
[Link]("CREATE TABLE IF NOT EXISTS users(id INT, name TEXT)")
for item in data:
[Link]("INSERT INTO users VALUES (?, ?)", (item['id'],
item['name']))
[Link]()
[Link]()
Design:
API → JSON → Database
Scalable and reusable
23(b) Strategy for missing values, duplicates, outliers (CR)
Missing values:
Drop if minimal
Impute (mean/median/mode)
Duplicates:
Use drop_duplicates()
Outliers:
Detect using IQR or Z-score
Remove or cap values
Strategy:
1. Identify issues
2. Apply appropriate cleaning
3. Validate results
24(a) Effectiveness of groupby (EV)
Advantages:
Efficient aggregation
Handles large datasets
Simplifies analysis
Example:
[Link]('category')['sales'].sum()
Evaluation:
Powerful for summarization
Reduces complexity
Limitation: Can be memory intensive
24(b) Pandas split-apply-combine workflow (CR)
Steps:
1. Split: Divide data into groups
2. Apply: Perform operations
3. Combine: Merge results
Example:
[Link]('dept')['salary'].mean()
Design Workflow:
Load data
Clean data
Group data
Apply functions
Combine and analyze
25(a) Visualization methods: categorical vs numerical (AN)
Categorical data:
Bar chart
Pie chart
Count plot
Numerical data:
Histogram
Line chart
Box plot
Analysis:
Categorical → compare groups
Numerical → show distribution/trends
Choosing correct chart improves interpretation
25(b) Visualization tools in Python (UN)
1. Matplotlib:
Basic plotting
import [Link] as plt
[Link]([1,2,3])
2. Seaborn:
Advanced statistical plots
import seaborn as sns
[Link](data)
3. Plotly:
Interactive visualization
Explanation:
Matplotlib → foundation
Seaborn → better aesthetics
Plotly → interactive dashboards