This also helps out my overall interest in nonprofit analytics, especially for smaller organizations. If we're gathering all this unstructured data, what if there's a wealth of information there that we're not yet putting together? And what if the excuse for that is, we can't afford the pricey packages out there now? For that reason, I wanted to make sure that what I demo'd is free, or at least exceedingly inexpensive (well...also because I'm a broke grad student. Either it's free--and preferably open source--or I keep on looking).
Anyone have experience with any of these packages I mentioned, or have a suggestion to add to the mix?
Also, I found a couple of articles about text analytics in general:
- Unstructured Data and the 80 Percent Rule - How much of our collected data is in unstructured text form? They estimate 80%, but it could be much higher. Can you imagine what's waiting for us to discover?
- Digging for Nuggets of Wisdom - Describes just the beginning of what we're finding possible.
- Taco Bell Takes Heat Over 'Drive-Thru Diet' Menu -The quote from this particularly made me laugh: "Prior to launch, posts were 73% positive, putting it ahead of beloved chains like Subway, Wendy's and Domino's. Words associated with the brand online were "love," "delicious," and "favorite." Postings are now 67% positive, putting Taco Bell behind White Castle, Blimpie and Arby's, which rank among the category's lower tier. Now three of the words most closely associated with Taco Bell and its campaign have been "fat," "stop," and "joke."
No comments:
Post a Comment