674 results for search: %EA%B4%91%EA%B3%A0%EB%AC%B8%EC%9D%98%E2%96%B7%E0%B4%A0%E2%9D%B6%E0%B4%A0%E3%85%A1%E2%9D%BD%E2%9D%BD%E2%9D%BC%E2%9D%BB%E3%85%A1%E2%9D%BD%E2%9D%BC%E2%9D%BC%E2%9D%BD%E2%96%B7%EA%B8%B0%EA%B3%84%EB%A9%B4%EA%B0%90%EC%84%B1%EB%A7%88%EC%82%AC%EC%A7%80%E3%81%8D%EA%B4%91%EA%B3%A0%E2%94%AE%EB%AC%B8%EC%9D%98%E2%86%82%EA%B8%B0%EA%B3%84%EB%A9%B4%E7%9C%98%EA%B0%90%EC%84%B1%EB%A7%88%EC%82%AC%EC%A7%80%E5%A6%B8emendatory/feed/rss2/press-release-chad-2010jan
Multiple Systems Estimation: Collection, Cleaning and Canonicalization of Data
<< Previous post: MSE: The Basics
Q3. What are the steps in an MSE analysis?
Q4. What does data collection look like in the human rights context? What kind of data do you collect?
Q5. [In depth] Do you include unnamed or anonymous victims in the matching process?
Q6. What do you mean by "cleaning" and "canonicalization?"
Q7. [In depth] What are some of the challenges of canonicalization? (more…)
Frequently Asked Questions
Multiple Systems Estimation
What is MSE?
What do you mean by statistical inference?
What is an overlap, and how do we know when lists overlap?
How does MSE find the total number of violations?
How was MSE originally developed?
How does the Benetech Human Rights Program use MSE?
1. What is MSE?
A: Multiple Systems Estimation, or MSE, is a family of techniques for statistical inference. MSE uses the overlaps between several incomplete lists of human rights violations to determine the total number of violations.
Return to Top
2. What do you mean by statistical inference?
A: ...
Reflections: Challenging Tasks and Meticulous Defenders
I have made it my personal objective to amplify HRDAG's message of being extra careful and scientifically rigorous with human rights data.
HRDAG Retreat 2016
What do you get when you bring seven statisticians, one quantitative political scientist, a writer, a computer scientist, and an administrator together for four days in a vacation rental on California’s Russian River? A lot of code, a technical paper and book chapter revised, another paper started, a great hike in the Redwoods, descriptions of food poisoning and crash landings in war zones, and a lot of talk about feelings.
Did I mention feelings? On the first evening of the annual retreat of the Human Rights Data Analysis Group, executive director Megan Price asked us to go around the room and share how we were feeling on arrival. The request ...
Data on Kosovo Migration
[popup citation="For migrations: Ball, Patrick. (2000). AAAS/ Human Rights Data Analysis Group database of migrations in Albania and Kosovo. For killings: Patrick Ball, Wendy Betts, Fritz Scheuren, Jana Dudukovich, and Jana Asher. (2002). AAAS/ABA-CEELI/Human Rights Data Analysis Group database of killings in Kosovo. For other data: Human Rights Data Analysis Group. (2002). Database of NATO airstrikes, geographic coding, and KLA activity in Kosovo."]
The data on migration from Kosovo are in seven files. All of the files are comma-delimited ASCII. The fields in each file are described below. For more information, see Policy or Panic, section A1, pp. ...
How much faith can we place in coronavirus antibody tests?
Given a positive test result, what is the probability that an individual has antibodies? This HRDAG-authored Granta article explains the science.
Data-driven crime prediction fails to erase human bias
Work by HRDAG researchers Kristian Lum and William Isaac is cited in this article about the Policing Project: “While this bias knows no color or socioeconomic class, Lum and her HRDAG colleague William Isaac demonstrate that it can lead to policing that unfairly targets minorities and those living in poorer neighborhoods.”
In Syria, Uncovering the Truth Behind a Number
Huffington Post Politics writer Matt Easton interviews Patrick Ball, executive director of HRDAG, about the latest enumeration of killings in Syria. As selection bias is increasing, it becomes harder to see it: we have the “appearance of perfect knowledge, when in fact the shape of that knowledge has not changed that much,” says Patrick. “Technology is not a substitute for science.”
HRDAG Retreat 2018
What follows is an elaborate criss-crossing of collaborations—retreat is a time to embrace the productivity that comes with being in the same room.
.outter-wrapper.feature {
background: #15795b;
}
.outter-wrapper.feature hr {
border-width: 0;
height: 30px;
}
.outter-wrapper.feature h4 {
/* height: 30px; */
border-width: 0;
}
.wrapper {
padding: 20px 0;
}
.branding-headline {
width: 100%;
font-size: 40px;
font-weight: 600;
padding-bottom: 20px;
color: #15795b;
line-height: 43.2px;
}
.border-line {
border-bottom: 1px solid #000;
margin: 20px 0;
}
.hed-dek-illo {
margin: 20px 0;
}
.illo {
width: 100%;
min-height: 200px;
}
.illo img {
margin: 0;
}
.blog-pages {
display: flex;
}
.blog-post {
flex: 0 0 ...
Multiple Systems Estimation: Stratification and Estimation
<< Previous post, MSE: The Matching Process
Q10. What is stratification?
Q11. [In depth] How do HRDAG analysts approach stratification, and why is it important?
Q12. How does MSE find the total number of violations?
Q13. [In depth] What are the assumptions of two-system MSE (capture-recapture)? Why are they not necessary with three or more systems?
Q14. What statistical model(s) does HRDAG typically use to calculate MSE estimates? (more…)
Reflections: Minding the Gap
How might we learn what we don’t know? HRDAG associate Christine Grillo hits the wayback machine and recalls her first exposure to People Against Bad Things, ideas about bias and correlation versus causation, and truth.
String matching for governorate information in unstructured text
code{white-space: pre;}
pre:not([class]) {
background-color: white;
}
h1 {
font-size: 34px;
}
h1.title {
font-size: 38px;
}
h2 {
font-size: 30px;
}
h3 {
font-size: 24px;
}
h4 {
font-size: 18px;
}
h5 {
font-size: 16px;
}
h6 {
font-size: 12px;
}
.table th:not([align]) {
text-align: left;
}
.main-container {
max-width: 940px;
margin-left: auto;
margin-right: auto;
}
code {
color: inherit;
background-color: rgba(0, 0, 0, 0.04);
}
img {
max-width:100%;
height: auto;
}
.tabbed-pane {
padding-top: 12px;
}
.html-widget {
margin-bottom: 20px;
}
button.code-foldin...
100 Women in AI Ethics
We live in very challenging times. The pervasiveness of bias in AI algorithms and autonomous “killer” robots looming on the horizon, all necessitate an open discussion and immediate action to address the perils of unchecked AI. The decisions we make today will determine the fate of future generations. Please follow these amazing women and support their work so we can make faster meaningful progress towards a world with safe, beneficial AI that will help and not hurt the future of humanity.
53. Kristian Lum @kldivergence
Media Contact
To speak with the researchers at HRDAG, please fill out the form below.
You can search our Press Room by keyword or by year.
HRDAG Retreat 2014
Ten data nerds gathered in a large hilltop beach house to analyze counts of killings from several war-torn countries. The time was January 16-20, 2014, the place was near San Francisco, the agenda was packed, and I was excited to be there.
Having defended my dissertation at Carnegie Mellon University just days before, I had often supposed that my thesis on a generalization of
log-linear models for capture-recapture might serve little other purpose than to fill a line on my curriculum vitae. This perception faded after a mid-2013 discussion with Patrick convinced me that HRDAG's data challenges could easily be the best match to my research ...
Courts and police departments are turning to AI to reduce bias, but some argue it’ll make the problem worse
Kristian Lum: “The historical over-policing of minority communities has led to a disproportionate number of crimes being recorded by the police in those locations. Historical over-policing is then passed through the algorithm to justify the over-policing of those communities.”
Privacy Policy
Mailing List Subscription
We use Mailchimp to help us keep track of community members who want to stay informed about what HRDAG is doing and thinking. If you self-subscribe to our list, we will never share your contact information. We will never subscribe anyone who does not explicitly agree to a subscription.
Over the course of a year, we mail quarterly letters and fundraising letters, as well as one or two updates as events demand. If, during the course of a fundraising campaign, you make a donation, we will do our best to remove you from the remainder of fundraising mailings that year. We may use your contact information to invite you to ...