Data Science Cheat Sheet for Business Leaders

Data Science Basics

Types of Data Science Building a Data Science Team
Descriptive Analytics (Business Intelligence): Get useful data in Your data team members require different skills for different purposes.
front of the right people in the form of dashboards, reports, and
Machine Learning
emails Data Engineer Data Analyst Data Scientist
Engineer
- Which customers have churned?
Store and maintain Visualize and Write production-level Build custom models to
- Which homes have sold in a given location, and do homes of a
data describe data code to predict with data drive business decisions
certain size sell more quickly?
SQL/Java/Scala/ SQL + BI Tools + Python/Java/R Python/R/SQL
Predictive Analytics (Machine Learning): Put data science Python Spreadsheets
models continuously into production
- Which customers may churn?
- How much will a home sell for, given its location and number of
rooms?
Data Science Team Organizational Models
Prescriptive Analytics (Decision Science): Use data to help a Centralized/isolated Embedded Hybrid
company make decisions The data team is the owner Data experts are Data experts sit with functional
- What should we do about the particular types of customers of data and answers dispersed across an teams and also report to the
requests from other teams organization and report Chief Data Scientist—so data
that are prone to churn?
to functional leaders is an organizational priority
- How should we market a home to sell quickly, given its location
and number of rooms? Data Engineering Design & Squad 1 Squad 2 Squad 3 Squad 1 Squad 2 Squad 3
Product

The Standard Data Science Workflow Data

Data Collection: Compile data from different sources and

1
store it for efficient access

Exploration and Visualization: Explore and visualize data

2
through dashboards

Experimentation and Prediction: The buzziest topic in data

3
science—machine learning!

www.datacamp.com/courses/data-science-for-business www.datacamp.com/groups/business
Exploration and Visualization Experimentation and Prediction
The type of dashboard you should use depends on what you’ll be using it for. Machine Learning
Common Dashboard Elements Machine learning is an application of artificial intelligence (AI) that builds
algorithms and statistical models to train data to address specific questions
Type What is it best for? Example without explicit instructions.

Supervised Machine Learning Unsupervised Machine Learning

Time series Tracking a value over time
Purpose Makes predictions from data Makes predictions by
with labels and features clustering data with no
labels into categories

Stacked bar chart Tracking composition over time Example Recommendation systems, email Image segmentation,
subject optimization, churn customer segmentation
prediction

Bar chart Categorical comparison

Popular Dashboard Tools

Spreadsheets
Special Topics in Machine Learning
BI Tools Customized Tools

Excel Power BI R Shiny Time Series Forecasting is a technique for predicting events through a
sequence of time and can capture seasonality or periodic events.
Sheets Tableau d3.js
Natural Language Processing (NLP) allows computers to process and analyze
Looker
large amounts of natural language data.
- Text as input data
- Word counts track the important words in a text
When You Should Request a Dashboard - Word embeddings create features that group similar words

When you’ll use it multiple times Deep Learning / Neural Networks enables Explainable AI is an emerging field in
unsupervised machine learning using data machine learning that applies AI such
that is unstructured or unlabeled. that results can be easily understood.

When you’ll need the information updated regularly

Highly accurate predictions Understandable by humans

Better for “What?” Better for “Why?"

When the request will always be the same

www.datacamp.com/courses/data-science-for-business www.datacamp.com/groups/business