McGraw Hill Notes- Ch7, Data Analytics
Data analytics attempts to analyze and extract appropriate data to help businesses in decision
making, help scientists/researchers to develop new products, help healthcare providers serve
the population, and more
KEY IDEAS
- What is data analytics, why is it used?
- What are the aspects of analytical thinking skills? There are 5
- What are trends from data analytics?
- What is a data ecosystem? How do you visualize one?
- What is DDDM? What are the THREE things it focuses on?
- Why do we use data visualization?
- How is SQL used? What is a Query and what is Syntax?
Process of Data Analytics
- Data analytics- the process of investigating raw data of various types to find trends and
correlations, and answer specifically made questions. THREE TYPES OF PROCESSES
- Descriptive analytics- works to uncover historical trends in data sets
- Answers questioned like “what happened?” or “what is occurring?”
- Deals with return on investment (ROI) and summaries of past events (ie. sales,
operational efficiency, impact of sales/marketing campaigns, and analysts of
survey data and the impact of a social media presence)
- Predictive analytics- focuses on understanding, predicting, and planning for future
events/outcomes
- Uses probability analysis techniques; data mining, statistical modeling, machine
learning, and deep learning to generate possible future outcomes in certain
conditions
- Used in e-commerce, cybersecurity, IT, and healthcare
- Prescriptive analytics- used to determine the best course of action. MOST
ADVANCED FORM OF DATA ANALYTICS
- Predict what, when, and why a situation will occur
- Ie. tracking fuel prices to predict increase and decreases, monitoring flu strains
and activity to determine outbreak areas, etc
Trends in Data Analytics
➔ Smarter, more responsible scalable AI - AI designed to be more responsive, smarter,
and scalable will help with short development times and better learning algorithms
- More orgs. are using AI because of the increased focus on Small Data and using
AI to operate with less data
- Big data- large chunks of structured/unstructured data
- Small data- small data sets that can impact decisions in the present
➔ Composable data and analytics- uses components from data sources/analytics
platforms to create a more user-friendly experience that's easier to find critical insights
- Creates new applications from existing analytics and data platforms to enable
more collaboration/insights
➔ Data fabric as a foundation- data fabric is a set of data services based on a specific
architecture that spans on premise and Cloud environments
, - Like a web that covers a large network across many locations
- Allows for more consistent and quicker data transfer/analysis
- Similar to how Internet is designed and operates
- Experts think the data fabric market will grow to 4.5 billion by 2026
➔ Data analytics as a core business function- today data analytics are treated as a
separate business function by a separate team
- Business leaders, including chief data officers (CDOs) are quickly realizing the
importance of data analytics and starting to treat it as a core business function
Uses of Data Analytics in Business
- Collection/analysis of data, better data visualization software, and ability to make more
informed decisions
★ Improved decision making- data analytics provides a whole view of customers and
business processes
○ Allows for continuous collection and analysis of data used to address problems
and make structured decisions
★ Improved customer service- data can be analyzed to find trends and create a more
customer-focused approach and strategy
★ Increased efficiency of production and operation processes- data analytics can e
used to assess production/operations processes, identify problems/opportunities and
provide potential solutions
★ Improved patient experiences/internal procedures in healthcare- data analytics can
be used to help patient experiences in treatment, billing, and overall satisfaction
○ Also used in pharmaceutical research/development
Ecosystems
- Data ecosystem- refers to the IT architecture/infrastructure, software applications,
programming languages, and storage technologies used for the collection, storage,
analysis, and interpretation of meaningful data
- They are unique, every organization generates/consumes data in different ways
Visualizing a data ecosystem: Data Project Lifecycle (Harvard University’s Data Science)
1. Sensing- Identification of meaningful data that should be collected by an organization
a. This stage determines the validity and usefulness of data
b. Areas of Data Used:
i. Organizational data sources (internal databases, spreadsheets, and other
sources of internal data)
ii. Data sources outside the organization
iii. Software used to collect data
iv. processes/systems used for data validation
2. Collection- gathering of data from determined validated data sources
a. Data can be collected manually or automatically captured from programs
b. Areas of Data Used:
i. Different programming languages to create programs to capture data
ii. The existing software to capture data
iii. Application programming interfaces (APIs) that collaborate with other
software applications to gather data
3. Wrangling- converting raw data into a user-friendly format
Data analytics attempts to analyze and extract appropriate data to help businesses in decision
making, help scientists/researchers to develop new products, help healthcare providers serve
the population, and more
KEY IDEAS
- What is data analytics, why is it used?
- What are the aspects of analytical thinking skills? There are 5
- What are trends from data analytics?
- What is a data ecosystem? How do you visualize one?
- What is DDDM? What are the THREE things it focuses on?
- Why do we use data visualization?
- How is SQL used? What is a Query and what is Syntax?
Process of Data Analytics
- Data analytics- the process of investigating raw data of various types to find trends and
correlations, and answer specifically made questions. THREE TYPES OF PROCESSES
- Descriptive analytics- works to uncover historical trends in data sets
- Answers questioned like “what happened?” or “what is occurring?”
- Deals with return on investment (ROI) and summaries of past events (ie. sales,
operational efficiency, impact of sales/marketing campaigns, and analysts of
survey data and the impact of a social media presence)
- Predictive analytics- focuses on understanding, predicting, and planning for future
events/outcomes
- Uses probability analysis techniques; data mining, statistical modeling, machine
learning, and deep learning to generate possible future outcomes in certain
conditions
- Used in e-commerce, cybersecurity, IT, and healthcare
- Prescriptive analytics- used to determine the best course of action. MOST
ADVANCED FORM OF DATA ANALYTICS
- Predict what, when, and why a situation will occur
- Ie. tracking fuel prices to predict increase and decreases, monitoring flu strains
and activity to determine outbreak areas, etc
Trends in Data Analytics
➔ Smarter, more responsible scalable AI - AI designed to be more responsive, smarter,
and scalable will help with short development times and better learning algorithms
- More orgs. are using AI because of the increased focus on Small Data and using
AI to operate with less data
- Big data- large chunks of structured/unstructured data
- Small data- small data sets that can impact decisions in the present
➔ Composable data and analytics- uses components from data sources/analytics
platforms to create a more user-friendly experience that's easier to find critical insights
- Creates new applications from existing analytics and data platforms to enable
more collaboration/insights
➔ Data fabric as a foundation- data fabric is a set of data services based on a specific
architecture that spans on premise and Cloud environments
, - Like a web that covers a large network across many locations
- Allows for more consistent and quicker data transfer/analysis
- Similar to how Internet is designed and operates
- Experts think the data fabric market will grow to 4.5 billion by 2026
➔ Data analytics as a core business function- today data analytics are treated as a
separate business function by a separate team
- Business leaders, including chief data officers (CDOs) are quickly realizing the
importance of data analytics and starting to treat it as a core business function
Uses of Data Analytics in Business
- Collection/analysis of data, better data visualization software, and ability to make more
informed decisions
★ Improved decision making- data analytics provides a whole view of customers and
business processes
○ Allows for continuous collection and analysis of data used to address problems
and make structured decisions
★ Improved customer service- data can be analyzed to find trends and create a more
customer-focused approach and strategy
★ Increased efficiency of production and operation processes- data analytics can e
used to assess production/operations processes, identify problems/opportunities and
provide potential solutions
★ Improved patient experiences/internal procedures in healthcare- data analytics can
be used to help patient experiences in treatment, billing, and overall satisfaction
○ Also used in pharmaceutical research/development
Ecosystems
- Data ecosystem- refers to the IT architecture/infrastructure, software applications,
programming languages, and storage technologies used for the collection, storage,
analysis, and interpretation of meaningful data
- They are unique, every organization generates/consumes data in different ways
Visualizing a data ecosystem: Data Project Lifecycle (Harvard University’s Data Science)
1. Sensing- Identification of meaningful data that should be collected by an organization
a. This stage determines the validity and usefulness of data
b. Areas of Data Used:
i. Organizational data sources (internal databases, spreadsheets, and other
sources of internal data)
ii. Data sources outside the organization
iii. Software used to collect data
iv. processes/systems used for data validation
2. Collection- gathering of data from determined validated data sources
a. Data can be collected manually or automatically captured from programs
b. Areas of Data Used:
i. Different programming languages to create programs to capture data
ii. The existing software to capture data
iii. Application programming interfaces (APIs) that collaborate with other
software applications to gather data
3. Wrangling- converting raw data into a user-friendly format
