HCDS (Fall 2017)

From CommunityData
Revision as of 17:36, 8 August 2017 by Jtmorgan (talk | contribs)
This page is a work in progress.
Last updated: 19:08, 15 July 2017 (EDT)
Human Centered Data Science
DATA 512 - Interdisciplinary Data Science Masters Program
Instructor
Jonathan T. Morgan
TA
Oliver Keyes
Course Website
We will use Canvas for announcements and turning in reading reflections, PAWS for turning in code, and Slack for Q&A and general discussion. All other course-related information will be linked on this page.
Course Description
Fundamental principles of data science and its human implications. Data ethics, data privacy, differential privacy, algorithmic bias, legal frameworks and intellectual property, provenance and reproducibility, data curation and preservation, user experience design and usability testing for big data, ethics of crowdwork, data communication and societal impacts of data science.

Goals and expectations

The format of the class will be a mix of lecture, discussion, analyzing data, in-class activities, short essay assignments, and programming exercises. Students will work in small groups. Instructors will provide guidance in completing the exercises each week.

Learning objectives

By the end of this course, students will be able to:

  • Analyze large and complex data effectively and ethically with an understanding of human, societal, and socio-technical contexts.
  • Develop algorithms that take into account the ethical, social, and legal considerations of large-scale data analysis.
  • Discuss and evaluate ethical, social and legal trade-offs of different data analysis, testing, curation, and sharing methods

Grading

Grades will be determined as follows:

  • 20% in-class work
  • 20% readings/reading groups
  • 60% assignments

Late assignments will not be accepted after the first week of class. In-class work and class participation cannot be made up. If you miss a class, you will receive a zero for the work done in class that day. Please do not ask the professor or TA what you missed during class; check the website or ask a classmate. Required posts to the class discussion board must be made before the due date or you will receive a zero for that work.

Final projects cannot be turned in late.


Schedule

HCDS (Fall 2017)/Schedule


Week 1: September 28

Day 1 plan

Day 1 slides

Course overview
What is data science? What is human centered? What is human centered data science?
Assignments due
  • fill out the pre-course survey
Agenda
  • Course overview & orientation
  • What do we mean by "data science?"
  • What do we mean by "human centered?"
  • How does human centered design relate to data science?


Readings assigned
Homework assigned
  • Reading reflection
Resources




Week 2: October 5

Day 2 plan

Day 2 slides

Ethical considerations in Data Science
privacy, informed consent and user treatment


Assignments due
  • Week 1 reading reflection
Agenda
  • Informed consent in the age of Data Science
  • Privacy
    • User expectations
    • Inferred information
    • Correlation
  • Anonymisation strategies


Readings assigned
  • Read: Markham, Annette and Buchanan, Elizabeth. Ethical Decision-Making and Internet Researchers. Association for Internet Research, 2012.
  • Read: Barocas, Solan and Nissenbaum, Helen. Big Data's End Run around Anonymity and Consent. In Privacy, Big Data, and the Public Good. 2014. (PDF on Canvas)
Homework assigned
  • Reading reflection
Resources




Week 3: October 12

Day 3 plan

Day 3 slides

Data provenance, preparation, and reproducibility
data curation, preservation, documentation, and archiving; best practices for open scientific research
Assignments due
  • Week 2 reading reflection
Agenda
  • Final project overview
  • Introduction to open research
  • Understanding data licensing and attribution
  • Supporting replicability and reproducibility
  • Making your research and data accessible
  • Working with Wikipedia datasets
  • Assignment 1 description


Readings assigned
Homework assigned
Examples of well-documented open research projects
Examples of not-so-well documented open research projects
Other resources





Week 4: October 19

Day 4 plan

Day 4 slides

Study design
understanding your data; framing research questions; planning your study


Assignments due
  • Reading reflection
  • A1: Data curation
Agenda
  • How Wikipedia works (and how it doesn't)
  • guest speaker: Morten Warnke-Wang, Wikimedia Foundation
  • Sources of bias in data science research
  • Sources of bias in Wikipedia data


Readings assigned


Homework assigned
  • Reading reflection
  • A2: Bias in data


Resources




Week 5: October 26

Day 5 plan

Day 5 slides

Machine learning
ethical AI, algorithmic transparency, societal implications of machine learning
Assignments due
  • Reading reflection
Agenda
  • Social implications of machine learning
  • Consequences of algorithmic bias
  • Sources of algorithmic bias
  • Addressing algorithmic bias
  • Auditing algorithms


Readings assigned
Homework assigned
  • Reading reflection
  • A3: Final project plan


Resources




Week 6: November 2

Day 6 plan

Day 6 slides

Mixed-methods research
Big data vs thick data; qualitative research in data science


Assignments due
  • Reading reflection
  • A2: Bias in data


Agenda
  • Guest speakers: Aaron Halfaker, Caroline Sinders (Wikimedia Foundation)
  • Mixed methods research
  • Ethnographic methods in data science
  • Project plan brainstorm/Q&A session


Readings assigned
Homework assigned
  • Reading reflection


Resources




Week 7: November 9

Day 7 plan

Human computation
ethics of crowdwork, crowdsourcing methodologies for analysis, design, and evaluation


Assignments due
  • Reading reflection
  • A3: Final project plan


Agenda
  • the role of qualitative research in human centered data science
  • scaling qualitative research through crowdsourcing
  • types of crowdwork
  • ethical and practical considerations for crowdwork
  • Introduction to assignment 4: Mechanical Turk ethnography


Readings assigned (read both, reflect on one)
Homework assigned
  • Reading reflection
  • A4: Crowdwork ethnography


Resources




Week 8: November 16

Day 8 plan

Day 8 slides

User experience and big data
user-centered design and evaluation of recommender systems; UI design for data science, collaborative visual analytics


Assignments due
  • Reading reflection
Agenda
  • HCD process in the design of data-driven applications
  • understanding user needs, user intent, and context of use in recommender system design
  • trust, empowerment, and seamful design
  • HCD in data analysis and visualization
  • final project lightning feedback sessions


Readings assigned
Homework assigned
  • Reading reflection


Resources




Week 9: November 23

Day 9 plan

Human-centered data science in the wild
community data science; data science for social good
Assignments due
  • Reading reflection
  • A4: Crowdwork ethnography
Agenda
  • NO CLASS - work on your own


Readings assigned
Homework assigned
  • Reading reflection
Resources




Week 10: November 30

Day 10 plan

Day 10 slides

Communicating methods, results, and implications
translating for non-data scientists


Assignments due
  • Reading reflection


Agenda
  • communicating about your research effectively and honestly to different audiences
  • publishing your research openly
  • disseminating your research
  • final project workshop


Readings assigned


Homework assigned
  • Reading reflection
  • A5: Final presentation
Resources




Week 11: December 7

Day 11 plan

Future of human centered data science
course wrap up, final presentations


Assignments due
  • Reading reflection
  • A5: Final presentation


Agenda
  • future directions of of human centered data science
  • final presentations


Readings assigned
  • none!
Homework assigned
  • none!
Resources
  • one




Week 12: Finals Week

  • NO CLASS
  • A6: FINAL PROJECT REPORT DUE BY 11:59PM on Sunday, December 10
  • LATE PROJECT SUBMISSIONS NOT ACCEPTED.


Assignments

HCDS (Fall 2017)/Assignments


Assignments are comprised of weekly in-class activities, weekly reading reflections, written assignments, and programming/data analysis assignments. Weekly in-class reading groups will discuss the assigned readings from the course and students are expected to have read the material in advance. In class activities each week are posted to Canvas and may require time outside of class to complete.

Unless otherwise noted, all assignments are due before 5pm on the following week's class.

Unless otherwise noted, all assignments are individual assignments.

Assignment timeline

Assignments due every week
  • In-class activities - 2 points (weekly): In-class activity output posted to Canvas (group or individual)
  • Reading reflections - 2 points (weekly): Reading reflections posted to Canvas (individual)


Scheduled assignments
  • A1 - 5 points (due Week 4): Data curation (programming/analysis)
  • A2 - 10 points (due Week 6): Sources of bias in data (programming/analysis)
  • A3 - 10 points (due Week 7): Final project plan (written)
  • A4 - 10 points (due Week 9): Crowdwork self-ethnography (written)
  • A5 - 10 points (due Week 11): Final project presentation (oral, written)
  • A6 - 15 points (due by 11:59pm on Sunday, December 10): Final project report (programming/analysis, written)

more information...


Readings

HCDS (Fall 2017)/Readings


coming soon


Policies

The following general policies apply to this course:

Respect
If there were only one policy allowed in a course syllabus, I would choose the word respect to represent our goals for a healthy and engaging educational environment. Treating each other respectfully, in the broadest sense and in all ways, is a necessary and probably sufficient condition for a successful experience together.
Attendance
Students are expected to attend class regularly.
Late Assignments
Late assignments will not be accepted. If your assignment is late, you will receive a zero score.
Participation
Active participation in class activities is one of the requirements of the course. You are expected to engage in group activities, class discussions, interactions with your peers, and constructive critiques as part of the course work. This will help you hone your communication and other professional skills.
Collaboration
Working in groups or on teams is an essential part of all data science disciplines. As part of this course, you will be asked to provide feedback of your peers' work.
Academic Integrity
Simply stated, academic integrity means that you are to do your own work in all of your classes, unless collaboration is part of an assignment as defined in the course. In any case, you must be responsible for citing and acknowledging outside sources of ideas in work you submit. Please be aware of the HCDE Department's and the UW's policies on this: HCDE Academic Conduct. These will be strictly enforced.
Assignment Quality
You are expected to produce work in all of the assignments that reflects the highest standards of professionalism. For written documents, this means proper spelling, grammar, and formatting.
Privacy
Students have the right for aspects of their personal life that they do not wish to share with others to remain private. Please respect that policy.
Accommodations
To request academic accommodations due to a disability, please contact Disabled Student Services: 448 Schmitz, 206-543-8924 (V/TTY). If you have a letter from DSS indicating that you have a disability which requires academic accommodations, please present the letter to me so you can discuss the accommodations you might need in the class.
Permissions
Unless you notify me otherwise, I will assume you will allow me to use samples from your work in this course in future instructional settings.
Disclaimer
This syllabus and all associated assignments, requirements, deadlines and procedures are subject to change.