MALLA REDDY ENGINEERING COLLEGE (AUTONOMOUS)

IV [Link] I Semester (MR 23) I Mid Question Bank 2025-26 (Objective)

Subject: Fundamentals of Data Science Branch: ECE

Name of the Faculty: Dr. Jyothula Sunil Kumar [Link]., Ph.D., MS in AI&DS (IIITB)

[Link] Questions Ans

MODULE – I (Introduction to Data Science)
Data Science is best described as:
a) A pure programming discipline
b) An interdisciplinary field using statistics, computing, and domain expertise to extract knowledge from
1 b
data
c) A subset of database management
d) A branch of hardware engineering
Which of the following is the PRIMARY goal of Data Science?
a) To write complex programs
2 b) To turn raw data into actionable insights and decisions b
c) To design computer networks
d) To manage IT infrastructure
Which of the following best defines 'Structured Data'?
a) Data with no predefined format, like audio files
3 b) Data organised in rows and columns, directly queryable with SQL b
c) Data stored in social media posts
d) Data that cannot be processed by computers
What percentage of the world's data is estimated to be UNSTRUCTURED?
4 d
a) 20% b) 40% c) 60% d) 80%
Which of the following is an example of SEMI-STRUCTURED data?
a) A relational database table
5 b) A CSV file c
c) A JSON file from a REST API
d) A raw sensor reading stored in a register
The CRISP-DM model stands for:
a) Cross Industry Standard Process for Data Management
6 b) Cross Industry Standard Process for Data Mining b
c) Computing Integrated Standard Process for Data Modelling
d) Comprehensive Industry Standard Protocol for Data Management
Which is the CORRECT order of phases in the CRISP-DM lifecycle?
a) Data Preparation → Business Understanding → Modelling → Evaluation
b) Business Understanding → Data Understanding → Data Preparation → Modelling → Evaluation →
7 b
Deployment
c) Modelling → Data Preparation → Evaluation → Deployment → Business Understanding
d) Data Collection → Analysis → Modelling → Deployment
A Data Scientist who primarily focuses on building, deploying, and scaling machine learning models in
production systems is best described as a:
8 c
a) Data Analyst b) Business Intelligence Analyst c) ML Engineer d) Database Administrator
Which of the following roles is responsible for building and maintaining the data pipelines (ETL
9 processes) that feed data into DS models? b
a) Data Scientist b) Data Engineer c) Business Analyst d) AI Researcher
10 The term 'Big Data' is commonly characterised by the three Vs. Which of the following is NOT one of d
them?
 a) Volume b) Velocity c) Veracity d) Visibility
Which of the following tools is MOST commonly used by Data Scientists for exploratory data analysis
11 and coding? b
a) Microsoft PowerPoint b) Jupyter Notebook c) Adobe Photoshop d) AutoCAD
Which Python library is PRIMARILY used for data manipulation and analysis in Data Science?
12 c
a) Matplotlib b) TensorFlow c) Pandas d) Scikit-learn
NumPy in Python is primarily used for:
a) Web development and routing
13 b) Numerical computations and array operations b
c) Natural Language Processing
d) Database querying
Matplotlib is used in Data Science for:
a) Data storage and retrieval
14 b) Data visualisation and plotting b
c) Machine learning model deployment
d) Statistical hypothesis testing
Which of the following best distinguishes Data Science from traditional statistics?
a) Data Science uses only smaller datasets
15 b) Data Science integrates programming, ML, and large-scale data to solve business problems b
c) Data Science ignores statistical methods
d) Statistics is only used in Data Science for visualisation
The 'V' that refers to the accuracy and reliability of Big Data is:
16 d
a) Volume b) Velocity c) Variety d) Veracity
Which of the following is an example of a PRIMARY data source?
a) Downloading a dataset from Kaggle
17 b) Using government census records c
c) Conducting a survey among students in your college
d) Reading a published research paper
Secondary data is BEST defined as:
a) Data collected first-hand by the researcher for the current study
18 b) Data previously collected by someone else for a different purpose b
c) Data with secondary importance to the research question
d) Data stored in the secondary memory of a computer
Which of the following is an example of a secondary data source?
a) Conducting interviews with hospital patients
19 b) Running an A/B test on a website c
c) Downloading historical weather data from [Link]
d) Observing customer behaviour in a retail store
The GIGO principle in Data Science stands for:
a) Gigabyte Input Gigabyte Output
20 b) Garbage In Garbage Out b
c) Global Input Generic Output
d) Graphical Input Graphical Output
Which of the following is NOT a common challenge in data collection?
21 c
a) Sampling bias b) Missing values c) Data encryption at rest d) Privacy and ethical concerns
Sampling bias in data collection occurs when:
a) The sample is too large to process
22 b) The collected sample does not accurately represent the target population b
c) The data is collected from too many sources
d) The sample is collected too quickly
23 A/B Testing is a method used in: b
a) Secondary data collection through scraping
b) Primary data collection — a controlled experiment comparing two versions
 c) Data storage optimisation
d) Model deployment and scaling
Which of the following libraries in Python is used for data VISUALISATION?
24 c
a) NumPy b) Pandas c) Seaborn d) SQLAlchemy
In the Data Science lifecycle, the step that immediately follows 'Data Collection' is:
25 c
a) Modelling b) Deployment c) Data Pre-Processing d) Business Understanding
Which of the following correctly describes the role of a 'Business Analyst' in a DS team?
a) Builds and trains ML models
26 b) Manages cloud infrastructure and data pipelines c
c) Bridges business goals and data teams; defines KPIs and requirements
d) Publishes research papers on new algorithms
Amazon's recommendation engine that accounts for approximately 35% of its revenue is an example of
27 Data Science application in: c
a) Healthcare b) Manufacturing c) Retail and E-commerce d) Transportation
Predictive maintenance in manufacturing uses Data Science to:
a) Hire more maintenance staff
28 b) Predict equipment failures before they occur, reducing downtime c
c) Manually inspect factory floors
d) Schedule government inspections
Which of the following BEST describes the role of Data Science in healthcare?
a) Managing hospital appointment scheduling only
29 b) Analysing patient records to detect diseases and predict outcomes b
c) Printing patient medical records
d) Billing and insurance processing
Real-time fraud detection in banking using ML models is an example of Data Science used in:
30 b
a) Agriculture b) Education c) Finance d) Transportation
Which of the following is the MOST important ethical concern when collecting personal data?
a) Using the fastest collection method
31 b) Ensuring informed consent and protecting individual privacy c
c) Maximising the volume of data collected
d) Storing data in multiple servers
India's DPDP Act 2023 (Digital Personal Data Protection Act) primarily governs:
32 a) Software development practices b) The collection, processing, and storage of personal data of Indian b
citizens c) Cybersecurity practices for government websites d) E-commerce taxation policies
Which of the following is an example of Data Science applied in Agriculture? a) Using ML to detect
33 crop diseases from smartphone images b) Building e-commerce websites for farmers c) Managing seed b
distribution warehouses d) Printing agricultural reports
The tool used to run Python code interactively in cells, widely used in DS education, is: a) PyCharm
34 a
IDE b) Visual Studio Code c) Jupyter Notebook / Google Colab d) MATLAB
Scikit-learn in Python is primarily used for: a) Data visualisation b) Database queries c) Machine
35 c
learning algorithms and model building d) Web scraping
Which of the following BEST describes 'Exploratory Data Analysis (EDA)'? a) The process of
36 deploying ML models to production b) Initial investigation of data using statistics and visualisation to c
discover patterns c) Collecting data from external sources d) Writing SQL queries to extract data
In Data Science, a 'Feature' refers to: a) A bug in the code that adds functionality b) An individual
37 measurable property or input variable used in ML models c) The output column of a dataset d) A b
visualisation tool in Pandas
The output variable that a machine learning model is trying to predict is called: a) Feature b) Input c)
38 b
Target variable (Label) d) Index
39 Which of the following is TRUE about the role of domain expertise in Data Science? a) It is unnecessary c
because algorithms learn everything from data b) It helps formulate the right questions and interpret
 results meaningfully c) It replaces the need for statistical analysis d) It is only needed for data collection,
not modelling
Web scraping is a secondary data collection technique that: a) Involves filling out online surveys
40 manually b) Extracts data from websites automatically using tools like BeautifulSoup c) Requires b
purchasing datasets from data providers d) Collects sensor data from IoT devices
Which Python library is used for making HTTP requests to APIs for data collection? a) Pandas b)
41 b
Requests c) NumPy d) Flask
The benefit of using SECONDARY data over PRIMARY data collection is: a) It is always more
42 accurate b) It is faster and cheaper to obtain c) It is always specific to the current research question d) It b
requires no quality evaluation
Which of the following CORRECTLY defines the 'Velocity' dimension of Big Data? a) The variety of
43 formats data comes in b) The truthfulness and accuracy of data c) The speed at which data is generated b
and must be processed d) The size of the dataset in terabytes
Terabytes, Petabytes, and Exabytes are units that relate to which Big Data 'V'? a) Velocity b) Variety c)
44 c
Volume d) Veracity
Which of the following is an example of Data Science applied in the EDUCATION sector? a) Using ML
45 to predict students at risk of dropping out b) Printing student hall tickets c) Managing library book c
inventories manually d) Building classroom furniture
A Data Science team working on a project at a hospital identifies that the patient dataset has missing
values in 15 columns. The FIRST step they should take is: a) Immediately fill all missing values with
46 a
zero b) Delete all rows with missing values c) Investigate WHY values are missing before deciding the
strategy d) Apply ML models directly to the incomplete dataset
Which of the following is NOT a typical responsibility of a Data Scientist? a) Building and evaluating
47 predictive models b) Communicating findings to non-technical stakeholders c) Installing physical c
network cables in the server room d) Performing exploratory data analysis
The concept that 'data has value like oil, but only when refined' is used to illustrate: a) That data must be
stored in oil-cooled servers b) That raw data alone is not useful — it must be processed to generate
48 c
insights c) That Data Science is an industry reserved for oil companies d) That data should be physically
stored underground
Which of the following Python commands would you use to import the Pandas library? a) import pandas
49 b
b) import pandas as pd c) from pd import pandas d) using pandas as pd
A student surveys 200 classmates about their study habits for a Data Science project. This is an example
50 of: a) Secondary data collection b) Web scraping c) Primary data collection d) Data from a government c
portal
MODULE – II (Data Pre-Processing)
Data Pre-Processing is defined as: a) The final step before reporting results b) The set of techniques to
51 convert raw, messy data into a clean, model-ready format c) The process of visualising data using charts b
d) The deployment of ML models to production servers
Which of the following is the CORRECT order of the 4 Data Pre-Processing steps? a) Transform →
52 Clean → Integrate → Reduce b) Clean → Integrate → Transform → Reduce c) Integrate → Clean → b
Reduce → Transform d) Reduce → Transform → Integrate → Clean
Why must Data CLEANING always be performed BEFORE Data TRANSFORMATION? a) Because
transformation algorithms are faster than cleaning algorithms b) Because outliers in raw data would
53 b
corrupt the scaling calculations during transformation c) Because cleaning requires more memory than
transformation d) Because transformation always creates duplicate records
Missing values in a dataset are BEST described as: a) Values that are equal to zero b) Values that are
54 d
negative c) Absent or null values in a dataset column d) Values greater than the column maximum
For a NUMERIC column with outliers, the MOST appropriate imputation strategy for missing values is:
55 a) Fill with the column mean b) Fill with the column median c) Fill with zero d) Fill with the column c
maximum
56 For a CATEGORICAL column such as 'gender' with missing values, the MOST appropriate imputation b
 strategy is: a) Fill with the median b) Fill with zero c) Fill with the column mode (most frequent value)
d) Fill with a random value
The Forward Fill (ffill) method for handling missing values is MOST appropriate for: a) Nominal
57 categorical data b) Time-series or ordered sequential data c) Binary classification targets d) High- b
dimensional feature vectors
MCAR (Missing Completely at Random) means: a) Data is missing due to the value being very high b)
58 Data is missing due to another observed variable c) Data is missing randomly, unrelated to any variable c
in the dataset d) Data is missing because the respondent refused to answer
Which type of missing data is the MOST dangerous for model building because it introduces systematic
59 bias? a) MCAR — Missing Completely at Random b) MAR — Missing at Random c) MNAR — b
Missing Not at Random d) Both MCAR and MAR equally
A student dataset has age values: [20, 21, 999, 19, 22, None]. What is the CORRECT sequence of
60 operations? a) Fill None with median → then remove 999 b) Replace 999 with NaN first (domain rule) d
→ THEN fill NaN with median c) Delete the entire column d) Replace both 999 and None with zero
In Python Pandas, the command to fill missing values in a column with the column median is: a)
61 df['col'].replace(df['col'].median()) b) df['col'].fillna(df['col'].median()) c) b
df['col'].dropna(df['col'].median()) d) df['col'].impute(df['col'].median())
Which Python command is used to REMOVE duplicate rows from a DataFrame? a)
62 c
df.remove_duplicates() b) df.delete_duplicates() c) df.drop_duplicates() d) df.clean_duplicates()
The IQR (Interquartile Range) method defines outliers as values that fall: a) Below the mean or above
63 the mode b) Below Q1 – 1.5×IQR or above Q3 + 1.5×IQR c) Below the 10th percentile or above the b
90th percentile d) More than 2 standard deviations from the mean
The Z-Score method flags a value as an outlier when: a) Its Z-score is exactly 0 b) The absolute value of
64 b
its Z-score is greater than 3 c) Its Z-score is negative d) Its Z-score is between -1 and +1
Winsorising (Clipping) as an outlier handling technique means: a) Removing all rows that contain
65 outliers b) Replacing outliers with the column mean c) Capping extreme values at defined percentile b
limits (e.g., 5th and 95th) d) Dividing the outlier value by 10
Data Integration refers to: a) Adding new columns to a single dataset b) Combining data from multiple
66 different sources into one unified dataset c) Splitting one large dataset into smaller files d) Converting d
data from one format to another
In Pandas, [Link](A, B, on='id', how='inner') returns: a) All rows from A and all rows from B with
67 NaN where no match b) All rows from A with NaN where B has no match c) Only rows that have c
matching 'id' values in BOTH A and B d) All rows from B with NaN where A has no match
A LEFT JOIN in Pandas (how='left') means: a) Only matching rows from both tables are kept b) ALL
68 rows from the LEFT table are kept; NaN added where no match in right table c) ALL rows from the b
RIGHT table are kept d) All rows from both tables are combined
[Link]([df1, df2], ignore_index=True) is used to: a) Join two DataFrames horizontally based on a
69 common key b) Stack two DataFrames VERTICALLY (append rows, same schema) c) Merge two c
DataFrames using outer join d) Perform a cross-product of two DataFrames
A common pitfall AFTER performing a merge/join operation is: a) All columns become strings b) The
70 DataFrame becomes too small c) Duplicate rows can be accidentally created — always run b
drop_duplicates() after d) The index always resets to zero
Min-Max Normalisation scales data to: a) Mean = 0 and Standard Deviation = 1 b) The range [0, 1] c)
71 c
The range [-1, +1] d) The original range of the data
The formula for Min-Max Normalisation is: a) x' = (x – mean) / std_dev b) x' = (x – min) / (max – min)
72 b
c) x' = x / max d) x' = (x – median) / IQR
Z-Score Standardisation scales data such that the result has: a) All values between 0 and 1 b) Mean = 0
73 b
and Standard Deviation = 1 c) All values becoming positive d) Maximum value = 100
The formula for Z-Score Standardisation is: a) x' = (x – min) / (max – min) b) x' = (x – mean) /
74 c
standard_deviation c) x' = x × 100 / max d) x' = (x – mode) / range
 Min-Max Normalisation is MOST appropriate for which type of machine learning algorithms? a)
75 Decision Trees and Random Forests b) K-Nearest Neighbours (KNN) and Neural Networks c) Naive c
Bayes classifiers d) Association rule mining
Z-Score Standardisation is MOST appropriate for which type of algorithms? a) K-Nearest Neighbours b)
76 Neural Networks with ReLU activation c) Support Vector Machines, Linear Regression, and PCA d) c
Decision Trees
Which of the following statements about Min-Max Normalisation is TRUE? a) It is not affected by
77 outliers b) If an outlier exists (e.g., age=999), it compresses all other values towards 0 c) It always b
produces mean=0 d) It makes all values negative
In sklearn, the class used for Min-Max Normalisation is: a) StandardScaler() b) NormalisationScaler() c)
78 c
MinMaxScaler() d) RangeScaler()
In sklearn, the class used for Z-Score Standardisation is: a) MinMaxScaler() b) StandardScaler() c)
79 b
ZScoreScaler() d) NormalScaler()
Label Encoding should ONLY be used for: a) Nominal (non-ordered) categorical variables like city,
80 branch, colour b) Ordinal (ordered) categorical variables like Low < Medium < High c) Any categorical b
variable regardless of ordering d) Numeric continuous variables
One-Hot Encoding is MOST appropriate for: a) Ordinal variables like grade: A > B > C > D b) Binary
81 variables like Yes/No c) Nominal variables with no natural order, like branch: CSE/ECE/MECH/IT d) c
Continuous numeric variables
The 'Dummy Variable Trap' in One-Hot Encoding refers to: a) Having too many categories b) Perfect
82 multicollinearity when all OHE columns are kept (one must be dropped) c) Encoding the same column b
twice d) Using OHE on numeric columns
In Pandas, the function used to perform One-Hot Encoding is: a) df.label_encode() b)
83 b
pd.get_dummies(df['col'], drop_first=True) c) [Link](df['col']) d) [Link](df['col'])
Feature Engineering refers to: a) Repairing broken features of a software system b) Creating NEW,
84 meaningful columns from existing ones using domain knowledge c) Removing irrelevant features from b
a dataset d) Converting features from one data type to another
If a student dataset has columns marks_m1, marks_m2, marks_m3, creating a new column 'avg_marks =
85 (m1+m2+m3)/3' is an example of: a) Data Cleaning b) Data Integration c) Feature Engineering d) Data c
Reduction
Data Reduction in pre-processing aims to: a) Collect more data from additional sources b) Remove
86 redundant or irrelevant features without losing predictive information c) Increase the number of columns b
in the dataset d) Convert all numeric data to text
The 'Curse of Dimensionality' refers to: a) Having too few samples to train a model b) Problems that
87 arise when a dataset has too many features, making models less accurate c) Missing values in b
dimensional columns d) Errors caused by incorrect unit conversion
If two features in a dataset have a Pearson correlation coefficient of 0.97, which action should be taken?
88 a) Keep both features as they are b) Drop one of them, as they carry redundant information c) Multiply b
them together to create a combined feature d) Apply PCA only to these two columns
The Variance Threshold technique in Data Reduction removes features where: a) The variance is very
89 high (above a set threshold) b) The variance is very low (near-constant features with little information) b
c) The mean is exactly zero d) The feature has more than 10 unique values
PCA (Principal Component Analysis) is used in Data Reduction to: a) Fill missing values in high-
90 dimensional data b) Combine correlated features into fewer uncorrelated principal components c) Detect b
duplicate rows in a dataset d) Encode categorical variables as binary columns
Which Python Pandas method shows the statistical summary (mean, std, min, max) of a DataFrame? a)
91 c
[Link]() b) [Link]() c) [Link]() d) [Link]()
[Link]().sum() in Pandas returns: a) The total number of rows in the DataFrame b) The number of
92 missing values per column c) The sum of all numeric values in the DataFrame d) The count of duplicate b
rows
 Which of the following is a correct statement about the pre-processing pipeline? a) Scaling should be
done before removing outliers b) The fitted scaler should be applied to BOTH training and test data
93 b
using the SAME fitted parameters c) OHE should always be performed before cleaning d) Feature
engineering must happen before data integration
In the context of pre-processing, 'standardising text data' means: a) Converting text to numbers b)
94 Making values like 'Male', 'male', 'M', 'MALE' consistent (e.g., all → 'Male') c) Removing all text b
columns from the dataset d) Translating text to a standard language
In Python, pd.to_numeric(df['col'], errors='coerce') converts: a) All values to integers b) Text that cannot
95 b
be converted to float to NaN, and valid strings to numbers c) All values to strings d) NaN values to zero
The KNN Imputer fills missing values using: a) The column mean across the entire dataset b) The
96 values of K most similar (nearest) rows c) The most frequent value in the column d) A random value b
from the column
Which of the following is the MOST important reason to apply the SAME fitted scaler to test data? a)
To make test data processing faster b) To prevent data leakage — the scaler should not 'see' or learn
97 b
from test data c) To ensure test data has the same number of rows as training data d) Because sklearn
requires the same scaler object for evaluation
What does df.drop_duplicates(subset=['student_id']) do? a) Removes all rows where student_id is null b)
98 Removes rows where student_id is duplicated, keeping the first occurrence c) Resets the student_id b
column to sequential numbers d) Removes the student_id column entirely
After applying One-Hot Encoding to a 'branch' column with 5 unique values (CSE, ECE, MECH, IT,
99 b
CIVIL), how many new binary columns are created if drop_first=True? a) 5 b) 4 c) 3 d) 6
Which of the following is a CORRECT statement about Feature Importance in Data Reduction? a)
Feature importance can only be determined by human domain experts b) Tree-based models like
100 b
Random Forest can rank features by their contribution to predictions c) All features must have equal
importance for the model to work d) Feature importance is only used in unsupervised learning
MODULE – III (Exploratory Data Analysis — EDA)
Exploratory Data Analysis (EDA) is best defined as: a) The process of training machine learning models
101 b) An initial investigation of data using statistics and visualisation to discover patterns c) The b
deployment phase of a Data Science project d) A method for collecting data from websites
Which of the following is NOT a goal of EDA? a) Identifying patterns and anomalies in data b)
102 Understanding the distribution of variables c) Building the final predictive model for deployment d) c
Discovering relationships between variables
103 The MEAN of the dataset [10, 20, 30, 40, 50] is: a) 25 b) 30 c) 35 d) 20 b
104 The MEDIAN of the dataset [5, 3, 8, 1, 9, 2, 7] is: a) 5 b) 4 c) 6 d) 7 a
105 The MODE of the dataset [2, 5, 3, 5, 8, 5, 2, 9] is: a) 2 b) 3 c) 5 d) 8 c
Standard Deviation measures: a) The central value of a dataset b) The most frequently occurring value c)
106 The spread or dispersion of data around the mean d) The difference between maximum and minimum c
values
Variance is related to Standard Deviation by: a) Variance = Standard Deviation × 2 b) Variance =
107 b
(Standard Deviation)² c) Variance = Standard Deviation / 2 d) Variance = √(Standard Deviation)
A HISTOGRAM is used to visualise: a) The relationship between two numeric variables b) The
108 distribution (frequency) of a single numeric variable c) Categorical data comparisons across groups d) b
Time-series trends
A SCATTER PLOT is used to visualise: a) The frequency distribution of one variable b) The
109 relationship and correlation between TWO numeric variables c) The proportion of each category in a b
dataset d) The distribution of categorical data
A BOX PLOT (Box-and-Whisker plot) is used to: a) Show the correlation between two continuous
110 variables b) Display the distribution, quartiles (Q1, Q3), median, and outliers of a numeric variable c) b
Compare proportions in a categorical variable d) Show trends over time
111 In a Box Plot, the line INSIDE the box represents: a) The Mean b) The Mode c) The Median (Q2) d) c
 The Range
112 The Pearson Correlation Coefficient (r) ranges from: a) 0 to 1 b) -1 to 0 c) -1 to +1 d) 0 to 100 c
If the Pearson Correlation Coefficient between two variables is 0.95, it indicates: a) A weak negative
113 c
correlation b) No correlation c) A strong positive correlation d) A perfect inverse relationship
A CORRELATION HEATMAP in Data Science is used to: a) Display the distribution of a single
114 variable b) Show the correlation coefficients between ALL pairs of numeric variables simultaneously c) b
Plot time-series data over months d) Compare categorical group averages
Which Python library is PRIMARILY used to create advanced statistical visualisations like violin plots,
115 c
heatmaps, and pair plots? a) NumPy b) Pandas c) Seaborn d) Requests
Skewness in a distribution describes: a) The height of the tallest bar in a histogram b) The asymmetry of
116 the data distribution around the mean c) The number of outliers in the dataset d) The correlation between b
two variables
A distribution is said to be POSITIVELY SKEWED (right-skewed) when:
117 a) Mean < Median < Mode b) Mean = Median = Mode c) Mean > Median > Mode d) Mode > Mean > c
Median
Kurtosis is a statistical measure that describes: a) The central value of a distribution b) The spread of
118 data around the median c) The 'peakedness' or tail-heaviness of a distribution compared to normal d) c
The number of modes in the dataset
A BAR CHART is MOST appropriate for: a) Showing the relationship between two continuous
119 variables b) Comparing values across different categories c) Displaying the distribution of a numeric b
variable d) Showing correlation between features
A PIE CHART is used to show: a) Trends over time b) The proportional composition of categories as
120 b
parts of a whole c) The distribution of a continuous variable d) Outliers in numeric data
Which of the following plots shows the distribution, probability density, and quartiles of a numeric
121 c
variable in a single visualisation? a) Bar chart b) Scatter plot c) Violin plot d) Pie chart
In Seaborn, which function creates a scatter plot matrix showing pairwise relationships between ALL
122 b
numeric columns? a) [Link]() b) [Link]() c) [Link]() d) [Link]()
The Five-Number Summary used in EDA consists of: a) Mean, Median, Mode, Range, Variance b)
123 Minimum, Q1, Median (Q2), Q3, Maximum c) Mean, Standard Deviation, Variance, Skewness, b
Kurtosis d) Min, Max, Mean, Median, Std
In a normally distributed dataset, approximately what percentage of data falls within ±1 standard
124 b
deviation of the mean? a) 50% b) 68% c) 90% d) 99.7%
A LINE CHART is MOST appropriate for visualising: a) The proportion of categories in a dataset b)
125 The frequency distribution of a numeric variable c) Trends and changes over time (time-series data) d) c
The correlation between two numeric variables

Signature of Faculty Signature of HOD