Ganesh Swami

What’s the best way to show a time series? A static plot only goes so far, and it’s actually very bad for presenting long timelines. Fortunately, designers have come up with creative uses of rich platforms such as Flash for interactive interfaces. All of these interfaces enable “zooming” to capture the level of detail.

Google Analytics

The new Google Analytics platform uses elements from Google Finance. You can click and drag a “window” to focus on a period of interest.

British History Timeline

BBC recently released an interactive Flash program to explore British history from the Neolithic (nothing much to see here) ages to today. This is very useful to our history discussions we have often in the lab.

Google Timeline View

At a conference last week, Google releases a couple of new interfaces to its engine. The timeline view plots the frequency of occurrence of search keywords. (I was talking to a friend about nanotechnology last week, and used this feature to show him when the hype peaked.)

The principle axes transform is a rigid registration technique for images. By taking the Singular Value Decomposition of the correlation matrix of fiduciary markers, the parameters of the transformation can be determined (this corresponds to three orientation and three translation in three dimensions.)

In information retrieval, this is known as latent semantic analysis. For example, if we have data linking customers and items they’ve bought, we can build a matrix with this information. Row-rank approximations of this matrix can be used to cluster customers. For data the size that Amazon has, the matrix is extremely sparse. For obvious reasons, this method doesn’t scale.

How does Amazon do it then? I didn’t know if the recommendations were done in real-time or updated offline like Google does with PageRank. That’s when some research took me to this publication: Amazon.com recommendations: item-to-item collaborative filtering. The publication has a decent review of existing methods: collaborative filtering, cluster models, and search-based methods and the reasons why they don’t scale for Amazon. More importantly, they aren’t fine-grained enough to recommend relevant items.

Amazon’s approach is a little different. Instead of grouping a user to a cluster of existing customers, they cluster items instead. For the details, read the paper (it also has pseudo-code.) This might sound obvious now, but it was pretty novel ten years back (a patent from 1998 by Amazon precedes the publication.) The algorithm scales independently of the number of customers and number of items in the product catalog. The computations of similar items is still expensive and done offline, but the retrieval can be done in real-time with high quality.

While browsing through various tutorials on typography, I came across the “right” way to use bullets. I’m not too sure about using this style though. Indenting bullets is more commonly seen in print and on the web, and attempting to use the right style will definitely stick out (for the wrong reasons.)