Live data from Hacker News

Portmanteau Generator

punchlinedesign.net

41–50 of 68 posts

Re: Portmanteau Generator

#43
post #29

The results are just incredible. The depth of the project is very deep. Case in point, chess+clock yields hourglasker (hourglass & Lasker). Lasker was a German chess champion, well known in chess circles but the average person would have no idea who he is. I wonder if this is using wikipedia or dbpedia to walk a graph and find words to stick together. That's the only way I could think of doing this.

Nope, the paper says it's just using the FastText word embedding: https://nips2018creativity.github.io/doc/entendrepreneur.pdf which is just a particularly well-done word2vec: https://arxiv.org/pdf/1712.09405.pdf 'Lasker' and 'chess' no doubt co-occur quite strongly in their Internet corpus.

On the other hand, it comes up with "Beagle" and "Labrador" as synonyms of "cat". Color me unimpressed.

This seems like something that has been done before, and the fact the paper has no references indicates to me that the author maybe didn't do background research. While [0] is different, I could have sworn I've seen a paper which discussed creating puns in this fashion.

[0] http://www.aclweb.org/anthology/W05-1614

Re: Portmanteau Generator

#44
post #43
post #29

Earlier quoted context omitted.

Nope, the paper says it's just using the FastText word embedding: https://nips2018creativity.github.io/doc/entendrepreneur.pdf which is just a particularly well-done word2vec: https://arxiv.org/pdf/1712.09405.pdf 'Lasker' and 'chess' no doubt co-occur quite strongly in their Internet corpus.

On the other hand, it comes up with "Beagle" and "Labrador" as synonyms of "cat". Color me unimpressed. This seems like something that has been done before, and the fact the paper has no references indicates to me that the author maybe didn't do background research. While [0] is different, I could have sworn I've seen a paper which discussed creating puns in this fashion. [0] http://www.aclweb.org/anthology/W05-1614

Maybe you saw https://llamasandmystegosaurus.blogspot.com/2017/05/alpha.ht... a year or two ago which does something similar in generating rhyming definitions, also using a word2vec approach.

Re: Portmanteau Generator

#46
post #43
post #29

Earlier quoted context omitted.

Nope, the paper says it's just using the FastText word embedding: https://nips2018creativity.github.io/doc/entendrepreneur.pdf which is just a particularly well-done word2vec: https://arxiv.org/pdf/1712.09405.pdf 'Lasker' and 'chess' no doubt co-occur quite strongly in their Internet corpus.

On the other hand, it comes up with "Beagle" and "Labrador" as synonyms of "cat". Color me unimpressed. This seems like something that has been done before, and the fact the paper has no references indicates to me that the author maybe didn't do background research. While [0] is different, I could have sworn I've seen a paper which discussed creating puns in this fashion. [0] http://www.aclweb.org/anthology/W05-1614

fasttext and word2vec only encode contextual similarity, they're not meant to generate synonyms.

Re: Portmanteau Generator

#48

Your generator keeps suggesting me to use the n-word (hard r) and I am not putting anything racist in. Can you please add some filters?

I would even go so far as to say he should just take that word out of his dictionary. There's never a time that it should serve it as a suggestion, IMO.

Re: Portmanteau Generator

#50
post #46
post #43

Earlier quoted context omitted.

On the other hand, it comes up with "Beagle" and "Labrador" as synonyms of "cat". Color me unimpressed. This seems like something that has been done before, and the fact the paper has no references indicates to me that the author maybe didn't do background research. While [0] is different, I could have sworn I've seen a paper which discussed creating puns in this fashion. [0] http://www.aclweb.org/anthology/W05-1614

fasttext and word2vec only encode contextual similarity, they're not meant to generate synonyms.

So, you agree with me then. You can't say "What do you call a sleepy cat? A /grumbea-gle/ (grumpy-beagle)!" It's a portmanteau, sure, but it is not really even close to the input. Using word2vec here is probably wrong, or should have different pruning for word variance. It's a flaw of the technique and choices of the author.
Post reply on HN