Learning Unit 4
Introduction
Data science is a broader field of study concerned with turning data into knowledge
Data science = “the art and science of acquiring knowledge through data.”
When we understand how to apply data to achieve our goals, we turn it into knowledge
Key concept:
Data = raw facts
Information = data processed into meaning
Knowledge = understanding how to apply information to achieve goals
Example: We used Excel charts to generate information, but data science connects this to
business goals
Data Science Process (5 Steps)
1. Formulate a question that you need an answer for
The process begins with identifying a business problem or question
Examples:
– “How can we improve customer satisfaction?”
– “How can we increase production plant performance?”
2. Collect data
Gather data needed to answer the question
Types of data sources:
– Primary data: collected specifically for the problem (e.g., surveys, interviews)
– Secondary data: pre-existing data (e.g., reports, databases)
Data cleaning is essential:
– Fix missing values, remove duplicates or invalid entries, ensure reliability
3. Explore the collected data (EDA)
Exploratory Data Analysis (EDA) = analyzing and investigating data sets, often using
visualization
Helps understand what the data represents
Output = information (but not yet actionable knowledge)
4. Create a model using the collected data
Apply machine learning or statistical techniques
Models can predict or classify
– Build model → evaluate model performance → improve where necessary
5. Create a visual representation of the results
Communicate results to business stakeholders clearly
Use visuals to present findings for better understanding
Once understood, stakeholders can implement actions based on results
Introduction
Data science is a broader field of study concerned with turning data into knowledge
Data science = “the art and science of acquiring knowledge through data.”
When we understand how to apply data to achieve our goals, we turn it into knowledge
Key concept:
Data = raw facts
Information = data processed into meaning
Knowledge = understanding how to apply information to achieve goals
Example: We used Excel charts to generate information, but data science connects this to
business goals
Data Science Process (5 Steps)
1. Formulate a question that you need an answer for
The process begins with identifying a business problem or question
Examples:
– “How can we improve customer satisfaction?”
– “How can we increase production plant performance?”
2. Collect data
Gather data needed to answer the question
Types of data sources:
– Primary data: collected specifically for the problem (e.g., surveys, interviews)
– Secondary data: pre-existing data (e.g., reports, databases)
Data cleaning is essential:
– Fix missing values, remove duplicates or invalid entries, ensure reliability
3. Explore the collected data (EDA)
Exploratory Data Analysis (EDA) = analyzing and investigating data sets, often using
visualization
Helps understand what the data represents
Output = information (but not yet actionable knowledge)
4. Create a model using the collected data
Apply machine learning or statistical techniques
Models can predict or classify
– Build model → evaluate model performance → improve where necessary
5. Create a visual representation of the results
Communicate results to business stakeholders clearly
Use visuals to present findings for better understanding
Once understood, stakeholders can implement actions based on results
