32 ms·
I've always wondered about doing this in non-flat spaces. Like if I add the "7100 miles west" vector to the "California" point, I get Turkmenistan. If I add "71
by imh 12y ago
I've always wondered about doing this in non-flat spaces. Like if I add the "7100 miles west" vector to the "California" point, I get Turkmenistan. If I add "7100 miles west" again, I get back near "California." Similarly, adding the "not" vector twice might get you back where you started in a word embedding. Anyone know if anyone is working on this? It could be tricky because "7100 miles west" lives in the tangent space to the space "California" lives in, but that in itself could be an interesting thing to study in the context of words.
- jeremysalwen 12y agoTake a look at some of the compositional models here under publications: http://www.socher.org/ http://www.socher.org/. Here is the demo webpage for the sentiment analysis system: http://nlp.stanford.edu/sentiment/ http://nlp.stanford.edu/sentiment/