Today was the first day of the two-day NSF Sponsored Symposium on Semantic Knowledge Discovery, Organization and Use at NYU.
Here are some highlights:
- Marti Hearst started us off with a discussion of tricks for statistical semantic knowledge discovery–namely, using “lots o’ text”, unambiguous cues, and “rewrite and verify”.
- Dekang Lin showed off the power of “lots o’ text” by showing how the Google n-gram data could be used to peform various semantic discovery tasks.
- Peter Turney argued that we need to combine symbolic representations for episodic information (i.e., what we obtain from information extraction) with spatial (i.e., vector space model) representations for semantic information.
There were a bunch of other talks that focused on the details of building and using semantic knowledge bases, but I’ll freely admit that I’m a bit of an outsider in this world. Nonetheless, I find the participation impressive in both quality and quantity.
I’ll post more notes tomorrow. And, if you’re in New York, I encourage you to attend tomorrow. They are letting people walk in, even they haven’t registered in advance.