« Why Do Airplanes Crash? Building an Open-Source Aircraft Sensor System | Main | Galois' Open-Source Projects on GitHub »

Tech Talk: An Analysis of Analysis

Galois is pleased to host the following tech talk. These talks are open to the interested public--please join us! (There is no need to pre-register for the talk.)

Please note the unusual date and time for this talk, it is on Friday, 11 May 2012, at 10:30am

title: An Analysis of Analysis

speaker: Charles Parker

time: Friday, 11 May 2012, 10:30am

Galois Inc.
421 SW 6th Ave. Suite 300,
Portland, OR, USA
(3rd floor of the Commonwealth building)


A basic problem in computer science is binary classification, in which an algorithm applies a binary label to data based on the presence or absence of some phenomenon. Problems of this type abound in areas as diverse as computational biology, multimedia indexing, and anomaly detection. Evaluating the performance of a binary labeling algorithm is itself a complex task, often based on a domain-dependent notion of the relative cost of "false positives" versus "false negatives". As these costs are often not available to researchers or engineers, a number of methods are used to provide a cost-independent analysis of performance. In this talk, I will examine a number of these methods both theoretically and experimentally. The presented results suggest a set of best practices for evaluating binary classification algorithms, while questioning whether a cost-independent analysis is even possible.

bio: Charles Parker received his Ph.D. in Computer Science in 2007 under Professor Prasad Tadepalli at Oregon State University. His thesis, "Structured Gradient Boosting", presented a gradient-based approach to structured prediction useful in information retrieval and planning domains. From 2007 to 2011, he worked for the Eastman Kodak Company on various problems in data mining, scanned document analysis, and consumer video indexing. He currently works for BigML, Inc., helping to develop a web-scale infrastructure and interface for machine learning. His work has appeared in ICML, AAAI, ICDM, and other notable venues.

Reader Comments

There are no comments for this journal entry. To create a new comment, use the form below.

PostPost a New Comment

Enter your information below to add a new comment.

My response is on my own website »
Author Email (optional):
Author URL (optional):
Some HTML allowed: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <code> <em> <i> <strike> <strong>