On Sat, Jul 18, 2009 at 4:02 PM, M. Edward (Ed) Borasky <[email protected]>wrote:
> > Netflix, even without the contributions of the contest teams, is doing > pretty well too. ;-) Different problem - they're aggregating votes, not trying to interpret language. Although it is certainly possible that some of the competitors are using third-party sources and linguistic analysis... I thought briefly about giving that a shot. > Man, it is so good to hear this from someone who's actually done it! > The other point, though, is that the "real thing", even traffic / > social network analysis, is compute-resource intensive and requires a > kind of programming knowledge that few have. So if something simple, > like emoticon counting, provides *some* clues about sentiment, it may > be worth doing. I'm not convinced, though, that it is worth doing. I'm not sure that's so true... there are a lot of tools out there that can be hooked together. The statistics and time series analytics call for some advanced knowledge, but I doubt if much of it is beyond a master's degree level. I found the harder parts to be figuring out what business problems can be solved, then packaging and presenting the data to people in a useful manner that also can be automated. There are a lot of graphs and visualizations, especially network visualizations, that work for the data at one point in time and become a mess when the data changes, so they are useless in an automated system. I was designing for executives who wanted everything summarized in a page... definitely a challenge. All the plumbing is hard to maintain, too, which is an argument for standards that would allow the pain to be shared. Nick
