Tag Archives: conference

Random Forest Verbal Autopsy Debut

I just got back from a very fun conference, which was the culmination of some very hard work, all on the Verbal Autopsy (which I’ve mentioned often here in the past).

In the end, we managed to produce machine learning methods that rival the ability of physicians. Forget Jeopardy, this is a meaningful victory for computers. Now Verbal Autopsy can scale up without pulling human doctors away from their work.

Oh, and the conference was in Bali, Indonesia. Yay global health!

I do have a Machine Learning question that has come out of this work, maybe one of you can help me. The thing that makes VA most different from the machine learning applications I have seen in the past is the large set of values the labels can take. For neonatal deaths, for which the set is smallest, we were hoping to make predictions out of 11 different causes, and we ended up thinking that maybe 5 causes is the most we could do. For adult deaths, we had 55 causes on our initial list. There are two standard approaches that I know for converting binary classifiers to multiclass classifiers, and I tried both. Random Forest can produce multiclass predictions directly, and I tried this, too. But the biggest single improvement to all of the methods I tried came from a post-processing step that I have not seen in the literature, and I hope someone can tell me what it is called, or at least what it reminds them of.

For any method that produces a score for each cause, what we ended up doing is generating a big table with scores for a collection of deaths (one row for each death) for all the causes on our cause list (one column for each cause). Then we calculated the rank of the scores down each column, i.e. was it the largest score seen for this cause in the dataset, second largest, etc., and then to predict the cause of a particular death, we looked across the row corresponding to that death and found the column with the best rank. This can be interpreted as a non-parametric transformation from scores into probabilities, but saying it that way doesn’t make it any clearer why it is a good idea. It is a good idea, though! I have verified that empirically.

So what have we been doing here?

7 Comments

Filed under TCS

Global Health Metrics and Evaluation (GHME) Conference 2011

The Global Health Metrics and Evaluation (GHME) Conference 2011 is something that might interest you. It’s in Seattle, from March 14-16, 2011. More from the conference website:

The Global Health Metrics & Evaluation (GHME) conference aims to bring together all the different disciplines involved in global health measurement and evaluation under one roof to share innovative tools and methods to get a better understanding of what the possibilities are in approaching population health measurement.

Who should come? People who are interested in cutting edge science, who want to learn from others in their fields and in other fields. Specifically, researchers, academic leaders, students, policymakers, non-governmental organizations, foundations, country offices of health statistics, and national and multi-national health organizations.

Comments Off

Filed under global health

Conference you should know about

This weekend marks the submission of my first “Global Health” paper. Congratulations to me! And many, many thanks to all the people who have worked with me to make it happen. I’ll go into details sometime in the future, first let me see how things go in the refereeing process.

While I was over-working on that business, I got an interesting Call-for-Papers forwarded from global health/AI researcher Emma Brunskill. The AAAI Spring Symposium on Artificial Intelligence for Development (AI-D) is an effort to build a community of people applying computer science and artificial intelligence in less-developed settings.

TCS people, don’t let the “AI” in their title turn you off. Eric Horvitz says that this is for all of us. Continue reading

Comments Off

Filed under global health, TCS