hyper-accurate estimation, classification, and quantification of unstructured data

Despite my initial trepidation, I participated in Fast Company’s The Influence Project.  These group contests aren’t very sophisticated in that they rely on the cult of personality as well as popularity to drive participation; which, to me feels…. well, like SPAM. My feeling or perhaps my hope, is that we will evolve as humans in this area and come up with smarter games.

Having said this, after locating and seeing my photo among the 32,955 others, I felt differently.  I saw myself as a small part of the world’s biggest team.  Awesome. That’s the power of visualization; an image can change how you think and feel about something.

Anyway, if you care to read between the lines, the following article has a little something for everyone interested in interpreting and visualizing data…

Unlike Crimson Hexagon, most social media monitoring companies rely on two common solutions: keywords and semantics.  Both, says Centurino, only offer non-specific positive or negative portraits of public opinion and are severely limited. Keywords analysis depends upon an expansive library of definitions (“It’s like Cirque du Soleil contortionism” he says); and a semantic approach (that is, the analysis of phrases and expressions) requires a language model that recognizes sarcasm, snark, abbreviations, and an endless amount of Web slang.

Harvard-Developed Tool Measures Real-Time Public Opinion on Social Media


Share

0 Responses to “hyper-accurate estimation, classification, and quantification of unstructured data”


Comments are currently closed.



Analytics Plugin created by Jake Ruston's Wordpress Plugins - Powered by Laptop Cases and r4.