Data Analysis by Algorithm

December 22, 2014

The folks at Google may have the answer for the dearth of skilled data analysts out there. Unfortunately for our continuing job crisis, that answer does not lie in (human) training programs. Google Research Blog discusses “Automatically Making Sense of Data.” Writers Keven Murphy and David Harper ask:

“What if one could automatically discover human-interpretable trends in data in an unsupervised way, and then summarize these trends in textual and/or visual form? To help make progress in this area, Professor Zoubin Ghahramani and his group at the University of Cambridge received a Google Focused Research Award in support of The Automatic Statistician project, which aims to build an ‘artificial intelligence for data science’.”

Trends in time-series data have thus far provided much fodder for the team’s research. The article details an example involving solar-irradiance levels over time, and discusses modeling the data using Gaussian-based statistical models. Murphy and Harper report on the Cambridge team’s progress:

“Prof Ghahramani’s group has developed an algorithm that can automatically discover a good kernel, by searching through an open-ended space of sums and products of kernels as well as other compositional operations. After model selection and fitting, the Automatic Statistician translates each kernel into a text description describing the main trends in the data in an easy-to-understand form.”

Naturally, the team is going on to work with other kinds of data. We wonder—have they tried it on Google Glass market projections?

There’s a simplified version available for demo at the project’s website, and an expanded version should be available early next year. See the write-up for the technical details.

Cynthia Murrell, December 22, 2014

Sponsored by ArnoldIT.com, developer of Augmentext

Comments

Comments are closed.

  • Archives

  • Recent Posts

  • Meta