Week 2 of the Data Science course covers discussions on big data and statistical sampling, required reading from introductory data science and machine learning texts, and an assignment to visually describe the Haberman dataset in R. The roadmap provides guidance on discussions, recommended reading, optional activities, and instructions for submitting the visual assignment by the deadline. It also previews upcoming weeks focusing on machine learning, data visualization, and data ethics.

1. Data Scientist Enablement
DSE 400 - Fast Track to Data Science
Week 2 Roadmap
Advanced Center of Excellence
Modern Renaissance Corporation
In Collaboration with SONO team and others
Content of this document is under Creative Commons Licence CC BY 4.0

2. Agenda
You can always find the latest version of this document at http://bit.ly/1dVHJwO
Week 1 Recap
Week 2 At a Glance
Discussions
Required Reading
Practice
Assignments and Submission
Looking ahead
References
Citation
Acknowledgement A strong will, a settled purpose, and an
invincible determination can accomplish almost
anything. - Thomas Fuller

3. During week 1 you were able to
Understand Data Science is and articulate what Data
Scientists do on day-to-day basis
Installed R and R-Studio
Explored UCI Machine Learning Repository
Import Housing Dataset into R
Explored SONO and participated in Discussions
DSE 400 - Week 1 Recap

4. Discussions:
Fuss about Big Data. Statistical sampling etc. Optional Q&A
Reading plan:
Read Chapters 4-7 from An Introduction to Data Science
R for Machine Learning by Allison Chung
Activities:
Play with spreadsheets, continue research on Data Viz. tools, connect with local groups etc.
Assignment 2:
Download Haberman dataset from UCI Machine Learning Repository into your R-Studio
environment and visually describe this dataset.
DSE 400 - Week 2 at a glance

5. Discussion 1: What’s all this fuss about Big Data? How would you go beyond
talking about 3 or 4 Vs of Big Data? Volume, Variety, Velocity, and Veracity
(by the way veracity means trustworthiness of this data). How about Value?
Do the people talk about it in the context of Big Data? Share your thoughts.
Discussion 2: “Statistics is defined as the discipline of using data samples to
support claims about populations.” Comments?
These discussions are required. These will be posted sequentially. If you have
access to SONO you are encouraged to participate in these discussions.
There will also be an Optional Q&A
For the sake of simplicity and ease of navigation, please do not create additional threads.
Social Engagement on SONO - Week 2
http://getsokno.com/redvinef/controllers/cell.php?user_knocell=1002

6. SONO or SOKNO (Social Knowledge platform) is chosen for the DSE program to enable
Social Engagement, Collaboration as well as Knowledge Dissemination which are all
important to an Open initiative like this.
We understand that many of you may be initially having navigational issues. To ease things,
here are some tweaks SONO team and the DSE community are developing, as we speak.
To enter a Knowledge Cell, login first then use the full url to enter right KC. For week 2 you
would use the following link
http://getsokno.com/redvinef/controllers/cell.php?user_knocell=1002
Weekly KCs DSE 400 Week 1, 2 ... etc. map to knocell numbers 1001, 1002 and so on on
these urls. Once you are in a KC click on threads to go to the current discussions. We
certainly appreciate your patience during this transitory phase.
SONO Tweaks

7. Read Chapters 4-7 from An Introduction to Data Science
Read R for Machine Learning by Allison Chung (Sections 1 to 3.4, pages 1-5 )
<Optional> Introduction to Probability and Statistics Using R (Chapters 1-3)
<Optional> Read Chapter 2-7 from Think Stats: Probability and Statistics for
Programmers
If you are unfamiliar with basic Statistical concepts or if you need a quick refresher
on this topic, please refer to Statistics Playlist by Khan Academy
Week 2 - Recommended Reading Plan
Data Scientists (n.): Person who is better at statistics than any software engineer and
better at software engineering than any statistician. - Josh Wills

8. <Practice> Given the following dataset, find manually the mean, median, mode, variance and standard
deviation for this population.
{ 3, 15, 17, 18, 20, 20, 12, 20, 20, 16, 17, 12, 4, 7, 15, 20, 12, 6, 1, 20 }
Also try using a spreadsheet (such as Excel or Google Spreadsheet) to find the above measures
for the same dataset.
<Practice> Math is Fun. Learn what Relative Frequency Distribution is. Try the example at the bottom
of the this page.
<Community Outreach> <Optional> Explore and connect with your local R Group (or Data
Scientist/Big Data groups) and check out their projects, talks and seminars that might interest you. Also
discuss with them how you can engage with them and help them out in their endeavors.
Activities

9. <Optional> If you are not fully happy with the statistical functionality of your familiar
spreadsheet package, download PSPP free statistical analysis tool from SourceForge and
play with it.
<Optional> <Advanced> Import Housing Data into R-Studio and describe it statistically.
You may need packages like pastecs which let you use stat.desc function.
<Optional> Register for Big Data in Motion This is a free online webinar scheduled for Jan
30, 2014, 1 PM EST. Attendance is optional but recommended.
Need more? Reach out to our Research Scholar Ms. Rachel Fleming
< Rachel@emodern.biz> and ask for more activities and challenges.
Activities - contd ...

10. Assignment 2 - Submission Required
Assignment: Download Haberman Survival dataset from UCI Machine Learning Repository. Import this
dataset into your R-Studio. Generate three graphic representations: Histogram, Scatter Plot and Box
Plot , as depicted above. Refer to R for Machine Learning by Allison Chung before you attempt this
assignment.
Image credit: R for Machine Learning by Allison Chung
<Help On Demand> You may
reach out to our Research
Scholar Ms. Rachel Fleming
<rachel@emodern.biz>
if you have any difficulties
with this assignment.

11. Submissions
Deadline Saturday, 11:59 PM your local time.
Mail Assignment 2 to <datascience400@gmail.com>
Submit a PDF document of the screenshots of your R-
Studio workspace showing the three visualizations
discussed. Use the naming convention: DSE 400 >
Assignment 2 > Your Full Name for your document.
No document links should be sent. Please add DSE 400
> Assignment 2 in the subject line.

12. Week 3 - 4 Intro to Machine Learning(ML) - Classification, Clustering, Prediction NaiveBayes,
Recommendations and Boosting algorithms . Refer to R for Machine Learning by Allison Chung
Watch Caltech Machine Learning Videos on Youtube
Week 5 Visualizations. Present your research Data Visualization Tools - A Comparative Study
Week 6 -7 Processing large data sets. Hadoop Ecosystem. Stream Computing etc.
Week 8 Ethics, Privacy and Building Data Products.
DSE 400 - Weeks 3-8 ahead

13. References, Resources and Additional Reading
An Introduction to Data Science
Think Stats: Probability and Statistics for Programmers
Statistics Playlist by Khan Academy
R for Machine Learning by Allison Chung
Introduction to Hypothesis Testing
Single Sample Hypothesis Testing Part 1 and Part 2
R for Beginners by Emmanuel Paradis
R - Reference Cards
Introduction to R Playlist (Video Collection) on Youtube
Caltech Machine Learning Playlist on Youtube
[MIT OCW] Prediction: Machine Learning and Statistics from MIT Sloan School of
Management,

14. Citation
The dataset titled Haberman's Survival Data used here for Assignment 2 comes from UCI
Machine Learning Repository
Donor for Haberman's Survival Data: Tjen-Sien Lim (limt@stat.wisc.edu). It was added UCI
Machine Learning Repository on March 4, 1999
R for Machine Learning by Allison Chung is recommended by MIT Course Prediction:
Machine Learning and Statistics from Sloan School of Management, It is adopted in DSE
400 as per OCW guidelines.
Content that appears as is on this document only, is under Creative Commons License CC
BY 4.0 This license may not necessarily apply to other material referenced here in this
document.

15. For More Information
Presentation deck for DSE 400 > Week 1 Roadmap can be found at http://bit.ly/1hC5wAV
Week 2 discussions take place during this week on SONO DSE 400 Week 2
<Help On Demand> You may reach out to our Research Scholar Ms. Rachel Fleming
<rachel@emodern.biz> if you have any difficulties with the assignments.
We welcome questions, thoughts and suggestions. Post these on SONO in the right
forum/discussion or write to us at <datascience400@gmail.com>
You can always find the latest version of this document at http://bit.ly/1dVHJwO

16. Fun@Work
Geographic distribution of clicks for Week 1 Roadmap

17. Fun@Work
Open Source Humor

18. We thank our community of committed and passionate volunteers, experts,
educators, innovators, benefactors, advisers, advocates, mentors and
supporters
We are also grateful to the outstanding support and encouragement from
SONO team as well as other organizations like MIT Sloan of Management,
IBM, HortonWorks, R-Project, Creative Commons, Open Courseware
Consortium, Stanford University, Caltech, O’Reilly Publications and Data
Science Central etc.
Acknowledgement

19. Thank You