BUSINESS ANALYTICS SPECIALIZATION

BUS 3.5.BA (R22): DATA VISUALIZATION & PREDICTIVE

ANALYSIS

Unit-I: Data visualization- Definition- key purposes - Types of data

visualizations: Tables- Pie charts and stacked bar charts- Line charts and area
charts- Histograms- Scatter plots- Heat maps- Tree maps; Advantages and
disadvantages of data visualization- importance of Data visualization and big
data- Data visualization examples. Most Popular Open Source Visualization
Libraries: D3.Js- Echarts- Vega- Deck.gl- Data visualization best practices.
Unit- II: Tools and software of data visualization: Best Data Visualization
Tools:Tableau - Looker -Zoho Analytics - Sisense - IBM Cognos Analytics
- Qlik Sense – Domo - Microsoft Power BIMN - Klipfolio - SAP Analytics
Cloud - Yellowfin – Whatagraph.
Unit-III: Predictive Analysis: Definition – Uses of Predictive Analysis -
Types of Predictive Analysis Models: Linear Regression - Random Forests
- Neural Networks - Predictive Analysis Tools: Rapid Miner Studio –
KNIME - IBM Predictive Analytics - SAP Predictive Analytics; Predictive
Analysis Examples: Financial Markets - Supply Chains
Unit-IV: Predictive Analysis Structure - Classification Algorithms - Predictive Analytics
Techniques - Algorithms in Predictive Analysis: Linear Regression - Logistic Regression -
Neural Network - Decision Trees - Naive Bayes; Applications of Predictive Analysis: Fraud
Detection - Health Care - Customer Targeting - Sales Forecasting - Assessing Risk -
Predictive Analytics vs Statistics - Predictive Analytics vs Data Mining

Unit-V: Predictive Forecasting – Forecasting - Predictive Forecasting - Key

Features of Predictive Forecasting - Forecasting vs Predictive Forecasting -
Predictive Forecasting vs Predictive Analytics - Top Predictive Forecasting
Tools: IBM Watson Studio - Power BI *
RapidMiner Studio – SAS - Use of Predictive Forecasting Tools: Demand
Prediction - Inventory Management - Identifying Growth Drivers -
Workforce Planning - Benefits of Predictive Forecasting.

Unit-I: Data visualization- Definition- key purposes - Types of data

visualizations: Tables- Pie charts and stacked bar charts- Line charts and area
charts- Histograms- Scatter plots- Heat maps- Tree maps; Advantages and
disadvantages of data visualization- importance of Data visualization and big
data- Data visualization examples. Most Popular Open Source Visualization
Libraries: D3.Js- Echarts- Vega- Deck.gl- Data visualization best practices

Here's an overview of data visualization:

Definition and Key Purposes

- Definition: Data visualization is the process of creating graphical

representations of data to better understand and communicate insights.
- Key Purposes:
1. To communicate complex data insights effectively
2. To facilitate data exploration and discovery
3. To support decision-making

Types of Data Visualizations

1. Tables: Used to display detailed data in a structured format

2. Pie Charts and Stacked Bar Charts: Used to show proportional
relationships between categories
3. Line Charts and Area Charts: Used to display trends and patterns over
time
4. Histograms: Used to show the distribution of continuous data
5. Scatter Plots: Used to visualize relationships between two continuous
variables
6. Heat Maps: Used to display relationships between two categorical
variables
7. Tree Maps: Used to display hierarchical data

Advantages and Disadvantages of Data Visualization

Advantages:
1. Improved understanding: Data visualization helps to communicate
complex data insights effectively
2. Faster decision-making: Data visualization facilitates data exploration
and discovery, supporting decision-making
3. Enhanced engagement: Data visualization can make data more engaging
and accessible to a wider audience

Disadvantages:

1. Misinterpretation: Data visualization can be misinterpreted if not

designed carefully
2. Information overload: Data visualization can be overwhelming if too
much data is displayed
3. Technical requirements: Data visualization requires technical expertise
and specialized tools

Importance of Data Visualization and Big Data

1. Big data insights: Data visualization helps to uncover insights from large
and complex datasets
2. Communicating complexity: Data visualization facilitates the
communication of complex data insights to non-technical stakeholders
3. Supporting decision-making: Data visualization supports decision-
making by providing actionable insights

Data Visualization Examples

1. Interactive dashboards: Used to display real-time data and support

decision-making
2. Storytelling with data: Used to communicate complex data insights
through narrative
3. Infographics: Used to display information in a visually engaging format
 Most Popular Open Source Visualization Libraries

1. D3.js: A JavaScript library for producing dynamic, interactive data

visualizations
2. ECharts: A comprehensive charting library for data visualization
3. Vega: A high-level visualization grammar for creating interactive data
visualizations
4. (link unavailable): A WebGL-based library for creating large-scale,
interactive data visualizations
5. Plotly: A popular library for creating interactive, web-based data
visualizations

Data Visualization Best Practices

1. Keep it simple: Avoid clutter and focus on communicating key insights

2. Use color effectively: Use color to draw attention and highlight important
information
3. Choose the right visualization: Select a visualization that effectively
communicates the data insights
4. Label and annotate: Provide clear labels and annotations to facilitate
understanding
5. Test and refine: Test the visualization with stakeholders and refine as needed

Unit- II: Tools and software of data visualization: Best Data Visualization Tools:Tableau -
Looker -Zoho Analytics - Sisense - IBM Cognos Analytics - Qlik Sense – Domo - Microsoft
Power BIMN - Klipfolio - SAP Analytics Cloud - Yellowfin – Whatagraph.

usiness Intelligence and Data Visualization Platforms

1. Tableau: A leading data visualization platform for creating interactive

dashboards.
2. Looker: A cloud-based business intelligence platform for data analysis
and visualization.
3. Qlik Sense: A business intelligence platform for data visualization and
analytics.
4. Sisense: A cloud-based business intelligence platform for data analysis
and visualization.
5. IBM Cognos Analytics: A business intelligence platform for data analysis
and visualization.

Cloud-Based Data Visualization Tools

1. Zoho Analytics: A cloud-based business intelligence platform for data

analysis and visualization.
2. Domo: A cloud-based business management platform for data integration
and visualization.
3. Klipfolio: A cloud-based dashboard platform for data visualization and
monitoring.
4. SAP Analytics Cloud: A cloud-based business intelligence platform for
data analysis and visualization.

Specialized Data Visualization Tools

1. Microsoft Power BI: A business analytics service for data visualization

and business intelligence.
2. Yellowfin: A business intelligence platform for data analysis and
visualization.
3. Whatagraph: A reporting and analytics platform for data visualization and
insights.

Key Features to Consider

1. Data Connectivity: Ability to connect to various data sources.

2. Data Visualization: Variety of visualization options and customization
capabilities.
3. Interactivity: Ability to interact with visualizations and drill down into
data.
 4. Collaboration: Ability to share visualizations and collaborate with others.
5. Security: Robust security features to protect sensitive data.
6. Scalability: Ability to handle large datasets and scale with growing needs.

Unit-III: Predictive Analysis: Definition – Uses of Predictive Analysis -

Types of Predictive Analysis Models: Linear Regression - Random Forests
- Neural Networks - Predictive Analysis Tools: Rapid Miner Studio –
KNIME - IBM Predictive Analytics - SAP Predictive Analytics; Predictive
Analysis Examples: Financial Markets - Supply Chains

Definition

Predictive analysis is a statistical technique used to predict future events or

behaviors based on historical data and statistical models.

Uses of Predictive Analysis

1. Risk Assessment: Identifying potential risks and opportunities.

2. Customer Behavior: Predicting customer behavior, such as churn or
purchase likelihood.
3. Financial Forecasting: Predicting financial outcomes, such as revenue or
stock prices.
4. Operational Optimization: Optimizing business processes, such as supply
chain management.

Types of Predictive Analysis Models

1. Linear Regression: A statistical model that predicts a continuous outcome

variable.
2. Random Forests: An ensemble learning method that combines multiple
decision trees.
3. Neural Networks: A machine learning model inspired by the human
brain.
 4. Decision Trees: A tree-like model that predicts an outcome based on a
series of decisions.

Predictive Analysis Tools

1. RapidMiner Studio: A data science platform for predictive analytics.

2. KNIME: An open-source data analytics platform.
3. IBM Predictive Analytics: A predictive analytics platform for business
users.
4. SAP Predictive Analytics: A predictive analytics platform for businesses.

Predictive Analysis Examples

1. Financial Markets: Predicting stock prices, credit risk, or portfolio

performance.
2. Supply Chains: Predicting demand, optimizing inventory, or identifying
potential disruptions.
3. Customer Relationship Management: Predicting customer churn,
purchase likelihood, or response to marketing campaigns.
4. Healthcare: Predicting patient outcomes, disease diagnosis, or treatment effectiveness.

Unit-IV: Predictive Analysis Structure - Classification Algorithms - Predictive Analytics

Techniques - Algorithms in Predictive Analysis: Linear Regression - Logistic Regression -
Neural Network - Decision Trees - Naive Bayes; Applications of Predictive Analysis: Fraud
Detection - Health Care - Customer Targeting - Sales Forecasting - Assessing Risk -
Predictive Analytics vs Statistics - Predictive Analytics vs Data Mining

Predictive Analysis Structure

1. Problem Definition: Define the problem or opportunity.

2. Data Collection: Gather relevant data.
3. Data Preparation: Clean, transform, and format data.
4. Model Development: Select and train a predictive model.
5. Model Evaluation: Evaluate the model's performance.
6. Deployment: Deploy the model in a production environment.
7. Monitoring and Maintenance: Monitor and maintain the model.

Classification Algorithms

1. Logistic Regression: A statistical model for binary classification.

2. Decision Trees: A tree-like model for classification and regression.
3. Random Forests: An ensemble learning method for classification and
regression.
4. Support Vector Machines (SVMs): A machine learning model for
classification and regression.
5. Neural Networks: A machine learning model inspired by the human
brain.

Predictive Analytics Techniques

1. Regression Analysis: A statistical method for predicting continuous

outcomes.
2. Time Series Analysis: A statistical method for forecasting future values.
3. Cluster Analysis: A statistical method for grouping similar data points.
4. Text Analytics: A method for extracting insights from unstructured text
data.
5. Machine Learning: A subset of artificial intelligence that enables
machines to learn from data.

Algorithms in Predictive Analysis

1. Linear Regression: A statistical model for predicting continuous

outcomes.
2. Logistic Regression: A statistical model for binary classification.
3. Neural Networks: A machine learning model inspired by the human
 brain.
4. Decision Trees: A tree-like model for classification and regression.
5. Naive Bayes: A probabilistic model for classification.

Applications of Predictive Analysis

1. Fraud Detection: Identifying potential fraudulent transactions.

2. Healthcare: Predicting patient outcomes, disease diagnosis, or treatment
effectiveness.
3. Customer Targeting: Identifying high-value customers or predicting
customer churn.
4. Sales Forecasting: Predicting future sales or revenue.
5. Assessing Risk: Identifying potential risks or opportunities.

Predictive Analytics vs Statistics

1. Focus: Predictive analytics focuses on predicting future outcomes, while

statistics focuses on understanding historical data.
2. Methods: Predictive analytics uses machine learning and data mining
techniques, while statistics uses traditional statistical methods.

Predictive Analytics vs Data Mining

1. Focus: Predictive analytics focuses on predicting future outcomes, while

data mining focuses on discovering patterns and relationships in data.
2. Methods: Predictive analytics uses machine learning and statistical techniques, while data
mining uses database querying and data visualization techniques.