September 2008
7 posts
A Bit of a Problem
Let’s talk about UI for a second, shall we? I do a lot of stuff on the web. I view a lot of content in the average day, I follow a lot of links. Frequently the sites I go to need me to sign in to see something or do something. That’s fine, nominally. The problem begins after I’ve signed in. Let’s take Flickr as an example. Some of the blogs I follow occasionally post...
Sep 18th
Moving On
I had just started college when 9/11 happened. I had, for all intents and purposes, just started thinking of myself as an adult. I remember that day very well. I woke up and it seemed like the world had ended. Remembering that my father had business in/near the Pentagon, I joined the hundreds trying to reach family and make sure they were okay. The day ended, but our problems surely...
Sep 11th
I was going to post a long cool thing on the __get__ method in python and how it works with functions, but it’s just too much of a pain in the ass to do that with tumblr (which powers this blog). Can someone out there tell me if there’s anything that can get some better formatting for code (like recognition of spaces and tabs?) that works for tumblr?
Sep 5th
Oh hey, by the way: As the last two posts probably indicate, I’m doing some less pretentious, more oh-hey-I’m-learning-about-this posts here. See, I’m still figuring this shit out, and I don’t much like acting like I know everything. So, there you are. If you’re interested in tracking them, they’re all tagged as ‘workbook’. How you use those tags...
Sep 2nd
1 tag
Support Vector Machines
This is old hat to anyone doing machine learning. Once upon a time I had a pretty good grasp on it myself. The trivial idea is that given two vectors representing two classes of something, you can divide the n-space that those two vectors are in roughly equally between them. Any new class of that something can then be identified by which side of that divider it falls into. The vectors in...
Sep 2nd
1 tag
Text Clustering and Latent Semantic Analysis
Work involves a lot of natural language processing and information retrieval. Or at least, it’s supposed to. With that in mind, I thought it would be worthwhile to do a bit of research/relearning on a major NLP technique, Latent Semantic Analysis (LSA). The basic idea of LSA is term clustering as a means of comparing documents. It’s pretty involved mathematics wise, and I...
Sep 2nd
Sep 2nd