Big Data Dialog

As we enter the age of Big Data, there is a great deal of good that we, as Data Scientists can do, but also the potential for great harm.  There is now a free online course on this topic.

You will find here several articles on the validity and fairness of our analyses, and on privacy preservation.  A high level overview of the ideas can be found in this keynote at IEEE Big Data.  All these thoughts can be succinctly summarized in terms of a code of ethics that all data scientists should live by.

Ask Me Anything about data science during a Reddit IAmA on December 7, 2016 at 10 a.m. EST

The Data Scientist's Code of Ethics

1. When collecting/analyzing data, I will not surprise the subject of the data.
   • Informed consent is the standard
   • Data Destruction pledge
2. I will own the outcome of my data analysis.
   • Address issues of algorithmic bias
   • Correct data errors as best as possible
   • Consider societal impact

Validity

Privacy

Fairness