DESIGN ANALYTICS

1.Data Reduction as a Data Pre-

processing
1. Improved Efficiency: Reduces
the amount of data to be
processed, speeding up analysis.
2. Memory Management: Helps in
fitting large datasets into
memory for analysis.
3. Enhanced Interpretability:
Simplifies data, making it easier
to visualize and understand.
4. Noise Reduction: Eliminates
redundant or irrelevant data,
improving model accuracy.
 5. Scalability: Facilitates the
handling of larger datasets by
compressing them into smaller,
manageable forms.

2. Identifying Data Quality and

Quality Measures
1. Accuracy: Ensures data values
are correct and reliable.
2. Completeness: Measures the
extent to which all required data
is present.
3. Consistency: Checks for
uniformity of data across
different sources and systems.
 4. Timeliness: Evaluates whether
data is up-to-date and available
when needed.
5. Uniqueness: Assesses the
presence of duplicates to
maintain data integrity.

3. Various Analytics Techniques

1. Descriptive Analytics:
Summarizes historical data to
understand past performance.
2. Inferential Analytics: Makes
predictions or generalizations
about a population based on a
sample.
 3. Predictive Analytics: Uses
historical data to forecast future
outcomes and trends.
4. Prescriptive Analytics:
Recommends specific actions
based on data-driven insights.
5. Exploratory Data Analysis
(EDA): Investigates datasets to
discover patterns and
relationships.

4. Data Imputation Techniques

1. Mean/Median/Mode
Imputation: Replaces missing
values with statistical measures
of central tendency.
2. K-Nearest Neighbors (KNN):
Estimates missing values based
on the values of similar data
points.
3. Regression Imputation:
Predicts missing values using
regression models based on
other variables.
4. Multiple Imputation: Creates
multiple datasets with different
imputed values to reflect
uncertainty.
5. Last Observation Carried
Forward (LOCF): Uses the last
 observed value to fill in missing
data for time series.

5. Need for Business Modeling

1. Strategic Alignment: Ensures
business processes support
overall organizational goals.
2. Operational Efficiency:
Identifies inefficiencies and
streamlines processes for better
performance.
3. Enhanced Decision-Making:
Provides a clear framework for
analyzing data and making
informed decisions.
 4. Risk Management: Helps in
identifying potential risks and
formulating mitigation strategies.
5. Communication Tool: Acts as a
visual representation to facilitate
communication among
stakeholders.

6. a) Apache Spark
1. Unified Engine: Supports batch
processing, streaming, machine
learning, and graph processing.
2. In-Memory Processing: Offers
faster data processing by storing
 data in memory rather than on
disk.
3. Scalability: Easily scales across
clusters of computers for
handling large datasets.
4. Support for Multiple
Languages: Compatible with
Python, Java, Scala, and R.
5. Extensive Libraries: Includes
built-in libraries for SQL,
machine learning (MLlib), and
graph processing (GraphX).
6. b) Cloudera Impala
1. SQL-Based Engine: Allows
users to run SQL queries on large
datasets in Hadoop.
 2. Low Latency: Designed for fast
query execution, enabling real-
time analytics.
3. Integration with Hadoop:
Works seamlessly with HDFS and
HBase for efficient data access.
4. Columnar Storage: Optimizes
performance by storing data in a
columnar format.
5. Compatibility: Supports
various BI tools for data
visualization and reporting.

7. What is Data? Handling Large

Collections of Data
1. Definition: Data is a collection
of facts, statistics, or information
that can be analyzed.
2. Types of Data: Includes
structured, unstructured, and
semi-structured data.
3. Storage Solutions: Utilizes
databases (SQL/NoSQL) for
effective data storage and
retrieval.
4. Distributed Processing:
Employs frameworks like Hadoop
and Spark for large-scale data
processing.
5. Data Pipeline Management:
Implements ETL (Extract,
 Transform, Load) processes to
handle data flows efficiently.

8. Constraints and Influences on

Data Architecture Design
1. Data Volume: High data
volumes require scalable storage
solutions.
2. Data Variety: Diverse data
types necessitate flexible and
adaptable architecture.
3. Data Velocity: Real-time data
processing needs impact design
choices.
 4. Compliance Regulations: Legal
and regulatory requirements
influence data governance and
security measures.
5. Technology Trends: Emerging
technologies can shape
architectural decisions for
efficiency and innovation.
9. Analytics Applications in
Various Business Domains
1. Healthcare: Analyzes patient
data for improved treatment
outcomes and operational
efficiency.
2. Finance: Risk assessment and
fraud detection using historical
 transaction data.
3. Retail: Customer behavior
analysis for targeted marketing
and inventory management.
4. Manufacturing: Predictive
maintenance to minimize
equipment downtime and
improve production efficiency.
5. Telecommunications: Churn
prediction to retain customers
and optimize service offerings.

10. Data Management and Steps in

Data Analysis
1. Definition: Data management
encompasses practices for
collecting, storing, and using
data securely and efficiently.
2. Data Collection: Gathering
relevant data from various
sources.
3. Data Cleaning: Identifying and
correcting errors and
inconsistencies in the dataset.
4. Data Exploration:
Understanding data
characteristics through
visualizations and summary
statistics.
 5. Data Interpretation: Analyzing
results to draw meaningful
conclusions and inform
decisions.

11. Applications of Data Modeling

in Business
1. Database Design: Structuring
data for efficient storage and
retrieval.
2. Business Process Mapping:
Visualizing processes to identify
bottlenecks and improve
workflows.
3. Data Integration: Aligning data
from multiple sources for
 comprehensive analysis.
4. Decision Support: Providing a
framework for analyzing data to
support strategic decisions.
5. Regulatory Compliance:
Ensuring data structures adhere
to industry regulations and
standards.

12. Differentiating SQL and NoSQL

1. Structure: SQL databases are
relational; NoSQL databases are
non-relational or semi-
structured.
2. Schema: SQL databases have
fixed schemas; NoSQL databases
allow flexible, dynamic schemas.
3. Query Language: SQL uses
Structured Query Language;
NoSQL uses various query
languages or APIs.
4. Data Integrity: SQL focuses on
ACID compliance; NoSQL often
prioritizes availability and
partition tolerance (CAP
theorem).
5. Scalability: SQL databases
typically scale vertically; NoSQL
databases are designed for
horizontal scaling.
13. Definition of Database and
Types
1. Database Definition: A
structured collection of data
stored electronically.
2. Relational Databases: Store
data in tables with predefined
schemas (e.g., MySQL, Oracle).
3. NoSQL Databases: Handle
unstructured or semi-structured
data; include document, key-
value, and graph databases (e.g.,
MongoDB, Cassandra).
4. Data Warehouses: Centralized
repositories designed for
analysis and reporting,
integrating data from multiple
sources.
5. Data Variables: Refer to
characteristics of the data, such
as categorical (qualitative) and
numerical (quantitative)
variables.