NYU Data Science newsletter – July 7, 2015

GROUP CURATION: N/A

… The growing variety of scenarios such as these has policy makers, practitioners and researchers seeking to understand the tensions and tradeoffs, risks and benefits of data-driven interventions. Indeed, the advances made in the data for development and humanitarian response arenas have informed related domains that grapple with similar issues – from technical issues over data visualization, predictive analytics, or encryption to social concerns like ethics, governance, and unintended consequences.

Yet from a human rights and human security perspective, knowledge gaps have emerged, particularly for situations that extend beyond crisis response or where aggregate data is inadequate.

You are a fully-equipped (or aspiring) data scientist and want to use your precious skills for solving problems that really itch the world? Welcome to the club. The good news is that there are many ways for data scientists to do good. However, the path is not always beaten and you might need to show some initiative. This article will give you some insight on how you can get involved, either through group meetings and events, as a volunteer or in paid positions.

Researchers say Google’s ad-targeting system sometimes makes troubling decisions based on data about gender and other personal characteristics.

… The Washington Post and the Guardian have developed their own methods for counting police-involved deaths in part because no other organization has precise tallies. The FBI, for its part, requests that law-enforcement agencies submit a count of police-involved deaths as part of its Uniform Crime Reports (UCR), the official data source on crime in the U.S. The FBI aggregates these counts and publishes estimates of justifiable homicides by police officers in its Supplementary Homicide Report. But the numbers are widely seen as incomplete because law-enforcement agencies contribute to the report voluntarily. Only about 2,700 out of 22,000 agencies recognized by the FBI contributed counts in 2013, the most recent year for which data is available. That year, the FBI tallied 461 homicides (its annual number has been around 400 for the last five years).

The chief technology officer of the Port Authority hopes to make more of the agency’s data available to outside developers, in part inspired by a recent PATH train-focused project at a hackathon and a data science school.

Three students at the New York City Data Science Academy analyzed PATH turnstile data to produce a visualization showing the circumstances leading to morning crowding on the trains from New Jersey to New York City.

This report on evaluating machine learning models arose out of a sense of need. The content was first published as a series of six technical posts on the Dato Machine Learning Blog. I was the editor of the blog, and I needed something to publish for the next day. Dato builds machine learning tools that help users build intelligent data products. In our conversations with the community, we sometimes ran into a confusion in terminology. For example, people would ask for cross validation as a feature, when what they really meant was hyperparameter tuning, a feature we already had. So, I thought, “Aha! I’ll just quickly explain what these concepts mean and point folks to the relevant sections in the user guide.”

I sat down to write a blog post to explain cross validation, hold-out data sets, and hyperparameter tuning. After the first two paragraphs, however, I realized that it would take a lot more than a single blog post.

This week, [Christopher] Bishop and Eric Horvitz, distinguished scientist and managing director of Microsoft Research’s Redmond lab, will host a panel of experts to discuss progress in artificial intelligence: myths, realities and aspirations. The panel is part of Microsoft Research’s annual Faculty Summit, and it will be streaming online for anyone who wants to listen.

The latest breakthroughs in artificial intelligence are the result of core advances in AI, including developments in machine learning, reasoning and perception, on a stage set by advances in multiple areas of computer science.

We are happy to announce the First Workshop on Data Systems for Interactive Analysis, to be held in conjunction with IEEE VIS 2015 in Chicago, October 26th, at the Palmer House Hilton in Chicago.

Deadline for Submissions: Friday, July 31