Applied Data

Science with R
Capstone project
<LEARNER’s Name>
<Date>
Outline
• Executive Summary
• Introduction
• Methodology
• Results
• Conclusion
• Appendix

2
Executive Summary
• Point1
• Point2
• Sub Point 1
• Sub Point 2
• Sub Point 3
• Point3
• Point4
• Point5

3
Introduction
• Point1
• Point2
• Point3
• Point4
• Sub Point1
• Sub Point2

4
Methodology
• Perform data collection
• Perform data wrangling
• Perform exploratory data analysis (EDA)
using SQL and visualization
• Perform predictive analysis using
regression models
• How to build the baseline model
• How to improve the baseline model
• Build a R Shiny dashboard app

5
Methodology

6
Data collection

• Describe how data sets were collected.

• You need to present your data collection

process use key phrases and flowcharts

• Add screenshots of Notebook code cell and cell

output used for OpenWeatherAPI and Webscrping
to the Appendix section for peer-review

7
Data wrangling

• Describe how data sets were processed

• You need to present your data wrangling process

using key phrases and flowcharts

• Add the screenshots of data wrangling code cell

and output for regular expressions, missing
values handling, generating indicator columns
to the Appendix section for peer-review

8
EDA with SQL

• Summarize performed SQL queries using bullet

points

• Add screenshots of all required SQL queries to

the Appendix section

9
EDA with data visualization

• Summarize what charts were plotted using bullet

points

• Add the screenshots of your ggplot code

snippets to the Appendix section

10
Predictive analysis

• Summarize how you built, evaluated, improved

and found the best performing model

• You need present your model development process

using key phrases and flowchart

11
Build a R Shiny dashboard

• Summarize what plots and interactions you built

into the dashboard using bullet points

12
Results
• Exploratory data analysis results

• Predictive analysis results

• A dashboard demo in screenshots

13
EDA with SQL

14
Busiest bike rental times

• Find dates and hours which had the most bike

rentals

• Present your query result with a short

explanation here

15
Hourly popularity and temperature
by seasons

• Find hourly popularity and temperature by

season

• Present your query result with a short

explanation here

16
Rental Seasonality

• Rental Seasonality

• Present your query result with a short

explanation here

17
Weather Seasonality

• Weather Seasonality

• Present your query result with a short

explanation here

18
Bike-sharing info in Seoul

• Find the total Bike count and city info for

Seoul

• Present your query result with a short

explanation here

19
Cities similar to Seoul

• Find all city names and coordinates with

comparable bike scale to Seoul's bike sharing
system

• Present your query result with a short

explanation here

20
EDA with Visualization

21
Bike rental Click icon to add picture
vs. Date
Show a scatter plot
of RENTED_BIKE_COUNT vs. DATE

Show the screenshot of the

scatter plot with explanations

22
Bike rental Click icon to add picture
vs. Datetime
Show the same plot of
the RENTED_BIKE_COUNT time
series, but now add HOURS as
the colour

Show the screenshot of the

scatter plot with explanations

23
Bike rental Click icon to add picture
histogram
Show a histogram overlaid with
a kernel density curve

Show the screenshot of the

histogram with explanations

24
Daily total
rainfall and Click icon to add picture
snowfall
Show a barchart calculating
the daily total rainfall and
snowfall

Show the screenshot of the box

plot with explanations

25
Predictive analysis

26
Ranked Click icon to add picture
coefficients
Show a screenshot of the
ranked coefficients bar chart
for the baseline model

Try to tell a story why some

variables are important while
some are not for predicting
bike-sharing demand

27
Model Click icon to add picture
evaluation
Built at least 5 different
models using polynomial terms,
interaction terms, and
regularizations

Visualize the refined models’

RMSE and R-squared using
grouped bar chart

28
Find the best performing model
• Select the best performing model with:
• RMSE must be less than 330
• R-squared must be larger than 0.72
• Shown a screenshot of the model performance

• Show its model formula here (RENTED_BIKE_COUNT ~

x1 + x2 + x3 ….)

• You could optionally present their final

coefficients here

29
Q-Q plot of the Click icon to add picture
best model
Plot the Q-Q plot of the best
model’s test results vs the
truths

30
Dashboard

31
<Dashboard screenshot 1>
• Replace <Dashboard screenshot 1> title with an
appropriate title

• Show the screenshot for cities’ max bike-

sharing prediction on a map

• Explain the important elements on the

screenshot

32
<Dashboard screenshot 2>
• Replace <Dashboard screenshot 2> title with an
appropriate title

• Show the screenshot when one specific city is

selected

• Explain the important elements on the

screenshot

33
<Dashboard screenshot 3>
• Replace <Dashboard screenshot 3> title with an
appropriate title

• Show the screenshot when another specific city

is selected

• Explain the important elements on the screenshot

34
CONCLUSION
• Point 1
• Point 2
• Point 3
• Point 4
•…

35
APPENDIX
• Include any relevant assets
like R code snippets, SQL
queries, charts, Notebook
outputs, or data sets that
you may have created during
this project

36