Business Intelligence

- Business intelligence (BI) includes a wide range of applications, practices, and technologies for
the extraction, transformation, integration, visualization, analysis, interpretation, and
presentation of data to support improved decision making.
- Many organizations use this data to build large collections of data called data warehouses,
data marts, and data lakes, for use in BI applications. Users, including employees,
customers, and authorized suppliers and business partners, may access the data and BI
applications via the Web or through organizational intranets and extranets—often using
mobile devices, such as smartphones and tablets. 

Benefits from BI and analytics

1. Detect Fraud
2. Improve Forecasting
3. Increase sales
4. Optimize operations
5. Reduce costs

Data Scientists

- Data scientists are individuals who combine strong business acumen, a deep understanding
of analytics, and a healthy appreciation of the limitations of their data, tools, and techniques
to deliver real improvements in decision making. 
- Analyze a scenario from different angles and make judgements
- Have a high educational background from Masters to Doctorate degree in analysis related
fields requiring math, statistics, computer science, MIS, etc.
- What if Analyses

Components Required for Effective BI and Analytics

- Data management: Obtaining, processing, managing, storage, output or dissemination of data

- Makes data valuable by being accessible, relevant, reliable, secure, simple, complete, etc.
- Data governance: Defining the roles of different positions in the organization regarding data
management
- Creative Data scientists: understands business processes, trends, and the technology and tools
required for effective business intelligence and analyses.

Business Intelligence and Analytics Tools

Spreadsheets

- Importing/inputting of data, performing operations, reporting the analysis through diagrams

- Excel Scenario Manager, Excel Solver Add-in: Finding optimal solutions to what-if problems

Reporting and Querying Tools

- Reporting and querying tools can present that data in an easy-to-understand fashion—via
formatted data, graphs, and charts.
 - Many of the reporting and querying tools enable end users to make their own data requests and
format the results without the need for additional help from the IT organization.

Data Visualization Tools

- Data visualization is the presentation of data in a pictorial or graphical format.

- reveal otherwise difficult-to-perceive patterns or relationships in the underlying
data.
- A word cloud is a visual depiction of a set of words that have been grouped together
because of the frequency of their occurrence.
- A conversion funnel is a graphical representation that summarizes the steps a consumer
takes in making the decision to buy your product and become a customer. It shows where
visitors to a Web site are dropping off the successful sales path.
- Conversion: consumer -> Customer

Online Analytical Processing

- Online analytical processing (OLAP) is a method to analyze multidimensional data from

many different perspectives.  It enables users to identify issues and opportunities as well as
perform trend analysis. 
- Databases built to support OLAP processing consist of data cubes that contain numeric
facts called measures, which are categorized by dimensions, such as time and geography. 

Drill-down analysis

- Drill-down analysis involves the interactive examination of high-level summary data in

increasing detail to gain insight into certain elements—sort of like slowly peeling off the
layers of an onion.
- Example:
1) Analysis of worldwide sales  2) Analysis in a specific country  Specific country at
specific month  sales of specific product line at specific country and month

Linear Regression

- Simple linear regression is a mathematical technique for predicting the value of a

dependent variable based on a single independent variable and the linear relationship
between the two.
- Linear regression consists of finding the best-fitting straight line (regression line) through
a set of observations of the dependent and independent variables. 

he regression line can be written as .

In this equation, the following are true:

- X is the value of the independent variable that is observed

- Y is the value of the dependent variable that is being predicted
- a is the value of Y when X is zero, or the Y intercept
- b is the slope of the regression line
- E is the error in predicting the value of Y, given a value of X
coefficient of determination, denoted r2 (and pronounced r squared) is a number that indicates
how well data fit a statistical model—sometimes simply a line or a curve. An r2 of 1 indicates that
the regression line perfectly fits the data, whereas an r2 of 0 indicates that the line does not fit the
data at all. An r2 of 0.92 means that 92 percent of the total variation in Y can be explained by the
linear relationship between X and Y as described by the regression equation. The other 8 percent of
total variation in Y remains unexplained. 
Data mining

- Data mining is a BI analytics tool used to explore large amounts of data for hidden patterns
to predict future trends and behaviors for use in decision making.
- Among the three most commonly used data mining techniques are
- association analysis (a specialized set of algorithms sorts through data and forms
statistical rules about relationships among the items),
- neural computing (historical data is examined for patterns that are then used to make
predictions), and
- case-based reasoning (historical if-then-else cases are used to recognize patterns).
- The Cross-Industry Process for Data Mining (CRISP-DM) is a six-phase structured
approach for the planning and execution of a data mining project 
Dashboard

Measures are metrics that track progress in executing chosen strategies to attain organizational
objectives and goals. These metrics are also called key performance indicators (KPIs) and consist
of a direction, measure, target, and time frame. 

 For a university. Increase (direction) the five-year graduation rate

for incoming freshman (measure) to at least 80 percent (target) starting
with the graduating class of 2022 (time frame).
 For a customer service department. Increase (direction) the
number of customer phone calls answered within the first four rings
(measure) to at least 90 percent (target) within the next three months
(time frame).
 For an HR organization. Reduce (direction) the number of
voluntary resignations and terminations for performance (measure) to
6 percent or less (target) for the 2018 fiscal year and subsequent years
(time frame).
A dashboard presents a set of KPIs about the state of a process at a specific point in time.
Dashboards provide rapid access to information, in an easy-to-interpret and concise manner, which
helps organizations run more effectively and efficiently.
Operational dashboards can be designed to draw data in real time from various sources, including
corporate databases and spreadsheets, so decision makers can make use of up-to-the-minute data.

Self-Service Analytics
Self-service analytics includes training, techniques, and processes that empower end users to
work independently to access data from approved sources to perform their own analyses using an
endorsed set of tools. 

Modern data management requires a true balancing act between enabling self-
service analysis and protecting sensitive business information.