DEPARTMENT OF DATA SCIENCE AND COMPUTER APPLICATIONS

FUNDAMENTALS OF BUSINESS ANALYTICS (HUM 4072)

IA 3 & IA 4

Submitted By:
Apurva Bhardwaj
(210968050)
 Introduction to PowerBI
01 course & learning
experience

Comparison of three
02 Business Analytics tools

CONTENTS Predictive and

03 Prescriptive Analytics

Privacy and Anonymity

04 Challenges on Hacking
Insider Threats
Course Completed-
“Microsoft - Data
Analysis and
Visualization with
Power BI”

3
 Introduction to Data Analysis and
Visualization using PowerBI
Course Overview:
The course titled "Data Analysis and Visualization with Power BI" covered the
fundamentals of data analytics and visualization using Microsoft Power BI.
It covered all the essential concepts of how to make a report, data modelling, and how
to generate insights using various visualization techniques.
Skills Acquired:
• Ability to build powerful visualizations, such as charts, graphs and maps
• Experience working with report-building capabilities and Power BI integration tools
• Increased knowledge of analytical techniques used in the interpretation of business
data and report creation.
Relevance to Industry:
With its ease of use, integration with other tools of Microsoft like Excel and Azure, and
cost-effectiveness, Power BI has found its wide industry application.
This is just the kind of solution business requirements call for scalable self-service BI for
decision-making.
Learning Experience of the course “Data Analysis
and Visualization using PowerBI”

I recently undertook the Microsoft course titled " Data

Analysis and Visualization using PowerBI ", which
focuses on the foundational aspects of using Power BI as a
business analytics tool. The course covered everything from
data preparation and visualization to creating interactive
reports and sharing insights. It provided a hands-on learning
experience, allowing me to work directly with real-world
datasets and explore how to generate meaningful reports
and dashboards. The course was structured to help both
beginners and experienced users leverage Power BI’s vast
capabilities effectively.
 Comparison of Business Analytics Tools
Three BA tools in this presentation are Microsoft PowerBI, Qlik Sense, Tableau.

1. Microsoft Power BI
PowerBI is a powerful business analytics tool by Microsoft which integrates with a variety of data sources and is
known for its user-friendly interface and affordability.
•Key Features:
• Data Visualization: It offers a wide range of customizable visualizations.
• Data Modelling: Has ability to utilize DAX queries for creating complex calculations and data manipulation.
• Collaboration: Seamlessly integrates with Microsoft Office Suite, facilitating easy sharing and collaboration.
•Use Cases:
• Ideal for small to large businesses looking for cost-effective data solutions.
• Effective for real-time data monitoring and reporting.

2. Qlik Sense
Qlik Sense is a powerful self-service data visualization and business intelligence tool designed for
interactive data exploration and discovery. It enables users to analyse vast amounts of data
effortlessly and gain insights through its intuitive interface.
•Key Features:
• Interactive Dashboards: Users can create interactive and highly customizable dashboards.
• Smart Search: Enables users to search across all data without predefined queries.
• Data Integration: Connects to various data sources for comprehensive data analysis.
•Use Cases:
• Suitable for organizations that require deep data exploration and intuitive analytics.
• Often used in industries like healthcare and finance for its robust data handling capabilities.

2. Tableau
Tableau is a leading data visualization and business intelligence tool that empowers users to transform raw data into interactive and
shareable dashboards. Renowned for its user-friendly interface, Tableau enables both technical and non-technical users to create
compelling visualizations quickly and efficiently.
•Key Features:
• Visual Analytics: Provides advanced visualization capabilities with drag-and-drop functionality.
• Collaboration: Strong emphasis on sharing dashboards via Tableau Server and Tableau Online.
• Data Blending: Allows users to combine data from multiple sources easily.
•Use Cases:
• Popular in various industries for complex data visualizations and real-time data analysis.
• Often preferred by data scientists and analysts for its powerful analytical capabilities.
 Comparison of Business Analytics Tools
Feature Power BI Tableau Qlik Sense
- User-friendly, intuitive - Interactive, highly - User-centric with a drag-and-
interface. customizable, but has a drop interface.
- Integrated with Microsoft suite steeper learning curve - Easy for beginners but has
(Excel, Azure, etc.), making it compared to Power BI. advanced options for power
Ease of Use ideal for businesses already - Requires some technical users.
using Microsoft products. expertise for advanced - Strong focus on simplicity in
- Drag-and-drop functionality, visualizations. creating dashboards and
making it easy for non-technical - Suited for data visualization visualizations.
users. experts.

Data Integration - Excellent for Microsoft - Offers a wide range of - Flexible data integration.
ecosystem integration (Azure, data connectors. - Seamlessly connects to
Excel, SQL Server). - Capable of integrating with multiple data sources, including
- Built-in connectors for hundreds almost any data source. SQL databases, cloud services,
of data sources like Google - Excellent for multi-source and APIs.
Analytics, Salesforce, and more. data blending and complex - Requires some technical
- Easily connects to both cloud data setups. knowledge for advanced data
and on-premise databases. - Highly customizable integration.
connections through APIs.
 Feature Power BI Tableau Qlik Sense

- Powerful visualizations with - Industry leader in creating - Strong focus on self-service

multiple templates. visually stunning and detailed visualizations.
- Suitable for simple and reports. - Highly interactive dashboards
complex reports. - Supports highly interactive with associative data models,
- Auto-refresh for live and customizable dashboards. allowing users to explore data
dashboards with real-time data. - Offers sophisticated visual freely.
Visualization & Reporting - Built-in AI and machine features and advanced - Slightly simpler visualizations
learning features for advanced charts. compared to Tableau.
insights.

- Free version available

- Free version available (Power - Free version available (Power
(Power BI Desktop).
BI Desktop). BI Desktop).
- Affordable pricing for the
- Affordable pricing for the Pro - Affordable pricing for the
Cost Pro version
version (₹830/user/month). Pro version (₹830/user/month).
(₹830/user/month).
- Scalable enterprise pricing for - Scalable enterprise pricing
- Scalable enterprise pricing
large organizations. for large organizations.
for large organizations.
 Feature Power BI Tableau Qlik Sense

- Real-time collaboration - Excellent sharing and - Qlik Sense Cloud allows easy
through Power BI Service. collaboration features through sharing and collaboration.
- Reports can be shared via the Tableau Online and Tableau - Dashboards can be published
cloud and accessed via mobile Server. and shared with stakeholders.
apps. - Dashboards can be - Supports mobile devices with
Collaboration and Sharing - Integrated with Microsoft embedded in websites or responsive design
Teams for seamless sharing. shared as interactive reports.
- Automatic data refresh ensures - Strong mobile integration.
up-to-date reports.

- Real-time collaboration - Excellent sharing and - Qlik Sense Cloud allows easy
through Power BI Service. collaboration features through sharing and collaboration.
- Reports can be shared via the Tableau Online and Tableau - Dashboards can be published
cloud and accessed via mobile Server. and shared with stakeholders.
apps. - Dashboards can be - Supports mobile devices with
- Integrated with Microsoft embedded in websites or responsive design
Collaboration and Sharing Teams for seamless sharing. shared as interactive reports.
- Automatic data refresh ensures - Strong mobile integration.
up-to-date reports.
 Feature Power BI Tableau Qlik Sense

- Built-in AI insights and machine - Built-in AI insights and - Built-in AI insights and
learning integration. machine learning integration. machine learning integration.
- DAX (Data Analysis - DAX (Data Analysis - DAX (Data Analysis
Advanced Analytics Features Expressions) allows complex Expressions) allows complex Expressions) allows complex
calculations. calculations. calculations.
- Predictive analytics powered - Predictive analytics powered - Predictive analytics powered
by Azure Machine Learning. by Azure Machine Learning. by Azure Machine Learning.

- Built-in security features with - Built-in security features with

- Built-in security features with
Microsoft Azure’s cloud Microsoft Azure’s cloud
Microsoft Azure’s cloud security.
security. security.
- Offers role-based access
- Offers role-based access - Offers role-based access
Security and Governance control and encryption.
control and encryption. control and encryption.
- Integrated with Active
- Integrated with Active - Integrated with Active
Directory for authentication and
Directory for authentication Directory for authentication
user management.
and user management. and user management.
 Predictive and Prescriptive Analytics
In today’s data-driven landscape, organizations leverage various analytical techniques to gain insights, make predictions, and
inform decisions. There are different types of data analytics with two prominent methodologies in this domain are Predictive
Analytics and Prescriptive Analytics. Understanding these analytics types is crucial for effective business decision-making.

PREDICTIVE ANALYTICS
Predictive Analytics involves using statistical algorithms, machine
learning techniques, and historical data to identify the likelihood
of future outcomes. It focuses on forecasting what might happen
based on past data trends.
•Key Techniques:
• Regression Analysis: Used to identify relationships
between variables and predict outcomes. For instance, a
company might use regression analysis to forecast sales
based on advertising spend.
• Time Series Analysis: Involves analysing data points
collected or recorded at specific time intervals to forecast
future values. This technique is widely used in finance for
stock price predictions.
• Machine Learning: Techniques like decision trees, neural
networks, and support vector machines are employed to
create models that can predict future trends based on 12
input data.
 •Applications:
• Sales Forecasting: Predicting future sales based on
historical performance and market trends.
• Customer Behaviour Prediction: Analysing customer
data to anticipate buying patterns and preferences.
• Risk Assessment: Identifying potential risks in financial
sectors, such as loan defaults, by predicting customer
creditworthiness.

PRESCRIPTIVE ANALYTICS
Prescriptive Analytics goes a step further by not only predicting future outcomes but also recommending actions to achieve
desired results. It utilizes optimization and simulation algorithms to suggest the best course of action.
Key Techniques:
• Optimization Algorithms: These mathematical models determine the most efficient allocation of resources to maximize
outcomes, such as profit or efficiency. For example, a delivery company can optimize routes to minimize transportation costs.
• Simulation Models: These models mimic the operation of a process or system over time, allowing organizations to test various
scenarios and their outcomes. For instance, a manufacturing firm might simulate production processes to determine the impact
of equipment downtime.
• Decision Analysis: This involves assessing different decision options and their potential consequences to guide organizations
toward the best choices. 13
Applications:
• Supply Chain Management: Optimizing inventory levels and
transportation routes to reduce costs and improve service levels.
• Marketing Strategies: Recommending targeted marketing
campaigns based on customer segmentation and behavior
predictions.
• Resource Allocation: Allocating human resources effectively in
project management based on predictive staffing models.

Aspect Predictive Analytics Prescriptive Analytics

Goal Forecast future events Recommend specific actions

Techniques Machine learning, regression Optimization, decision trees
Example Sales trend predictions Inventory management solutions

14
 Privacy and Anonymity Challenges in
Managing Insider Threats
• Insider Threats: These occur when individuals within an organization, such as employees, contractors, or business partners,
exploit their authorized access to harm the organization. They may steal data, sabotage systems, or leak sensitive
information.
• Privacy and Anonymity Challenges:
• Anonymity vs. Security: Organizations face a dilemma in maintaining employee privacy while ensuring adequate
security measures are in place to detect and prevent insider threats. Striking the right balance is crucial.
• Monitoring Risks: Monitoring employee behaviour and access to sensitive data is essential for detecting threats, but
excessive surveillance can infringe on personal privacy and create a sense of distrust among employees.
• Data Collection and Anonymization: When monitoring data access, it's important to anonymize or pseudonymize
employee information to protect their privacy. However, this can limit the ability to detect suspicious patterns tied to
specific individuals.
• Common Insider Threat Scenarios:
• Malicious Insiders: Deliberate actions by individuals to cause harm, such as stealing intellectual property or customer
data.
• Negligent Insiders: Unintentional actions by employees, like falling victim to phishing scams or accidentally sharing
confidential information.
• Compromised Insiders: External attackers who gain access to internal systems by exploiting legitimate employee
credentials. 15
• Impact of Insider Threats:
• Financial Loss: Data breaches caused by insiders often result in significant financial costs due to legal fees, regulatory
fines, and loss of customer trust.
• Reputation Damage: Insider threats can severely damage an organization's reputation, especially if sensitive customer
information is compromised.
• Intellectual Property Theft: Insider threats pose a major risk for organizations dealing with patents, trade secrets, or
proprietary technology.
• Real-World Examples:
• Edward Snowden Case (2013): A former NSA contractor who leaked classified information, highlighting the challenges of
detecting trusted insiders who misuse access.
• Target Data Breach (2013): Compromised insider credentials led to a data breach affecting millions of customers,
illustrating how external attackers can exploit insider weaknesses.

16
Strategies for Detecting and Preventing Insider Threats
• Preventive Techniques:
• User Behaviour Analytics (UBA):
• UBA involves monitoring the digital behaviour of employees, looking for unusual activities that could indicate insider
threats. It uses machine learning algorithms to detect anomalies, such as accessing files outside normal work hours,
unauthorized data downloads, or unusual patterns of email communication.
• Example: An employee downloading large amounts of sensitive data after submitting their resignation could trigger an
alert.
• Access Control and Least Privilege Principle:
• This strategy limits access to sensitive information based on job roles, ensuring that employees can only access the
data necessary for their job functions. By implementing least privilege, organizations reduce the risk of accidental or
intentional misuse of data.
• Example: A marketing employee wouldn’t have access to financial or HR systems unless explicitly required for their role.
• Multi-Factor Authentication (MFA):
• MFA adds an extra layer of security by requiring users to provide two or more verification methods before accessing
sensitive data or systems. This reduces the likelihood of compromised credentials leading to unauthorized access.
• Example: Requiring a second layer of authentication, such as a mobile token, even after entering a password.

17
• Technological Tools for Detection:
• Artificial Intelligence and Machine Learning:
• AI-based systems can predict insider threats by analysing employee behaviour in real time. These systems detect subtle
signs of insider threats by correlating multiple factors (e.g., emotional changes seen in communication patterns,
abnormal system usage).
• Example: An AI system could detect patterns of frustration or dissatisfaction in internal communication, flagging it as a
potential precursor to malicious actions.
• Balancing Privacy and Security:
• The Privacy Challenge:
• While it's essential to monitor employee behaviour to detect threats, organizations must balance this with privacy laws
and ensure they aren’t violating employee rights. Intrusive surveillance can erode trust, leading to morale issues.
• Anonymization of Monitoring Data: One solution is to anonymize or pseudonymize data to protect individual identities
while still tracking suspicious activities.
• Ethical Considerations:
• Companies must establish transparent policies on how employee data is monitored and clearly communicate these
policies. Employees should be aware of what is monitored and why, without feeling that their personal privacy is
invaded.
• Example: Letting employees know their access to certain systems is monitored, but personal emails and private
communications are not tracked.
THANK YOU