Data Science for Social Good is a 2-unit DeCal designed to give students an opportunity to analyze ethical issues in data science and gain hands on experience using data science for social good. The course is split into different modules, each of which will focus on various case studies as well as relevant data science and machine learning methods. Additionally, we will address relevant ethical issues that arise from the case studies and methods discussed, as well as strategies to recognize and tackle ethical challenges. Technical material will review relevant statistical/data science/machine learning techniques and explore ways in which incorrect applications of these methods (or incorrect conclusions drawn from these methods) can result in negative consequences. Additionally, students will have the opportunity to dive deeper in the material and apply what they’ve learned in class through various small assignments (see syllabus for more details). The course will conclude with a final project where students will have the opportunity to apply what they have learned throughout the semester.

Prerequisites:
The ideal student will have taken DS100 (or be concurrently enrolled). Students should have familiarity with pandas and some basic machine learning exposure either through a course (such as EE16AB, DS100, CS188, CS189, relevant MOOC) or research/work experience. Specifically, students should have been exposed to basic regression and classification techniques and basic statistics. Students that have not taken DS100 but have programming experience equivalent of that gained in CS61A and some basic machine learning exposure will be well prepared. Students who do not meet these prerequisites but feel they have sufficient programming and statistical knowledge are welcome to enroll and encouraged to speak to the facilitators about their background.

Students will be given permission codes at the first class meeting on Sept 4. If student interest exceeds available spots, permission codes will be distributed via lottery.