We describe a convolutional neural network that learns feature representations for short textual posts using hashtags as a supervised signal. The proposed approach is trained on up to 5.5 billion words predicting 100,000 possible hashtags. As well as strong performance on the hashtag prediction task itself, we show that its learned representation of text (ignoring the hash-tag labels) is useful for other tasks as well. To that end, we present results on a document recommendation task, where it also outperforms a number of baselines.
%0 Conference Paper
%1 conf/emnlp/WestonCA14
%A Weston, Jason
%A Chopra, Sumit
%A Adams, Keith
%B EMNLP
%D 2014
%E Moschitti, Alessandro
%E Pang, Bo
%E Daelemans, Walter
%I ACL
%K cnn embedding hashtag mining semantic tagspace text
%P 1822-1827
%T #TagSpace: Semantic Embeddings from Hashtags.
%U http://dblp.uni-trier.de/db/conf/emnlp/emnlp2014.html#WestonCA14
%X We describe a convolutional neural network that learns feature representations for short textual posts using hashtags as a supervised signal. The proposed approach is trained on up to 5.5 billion words predicting 100,000 possible hashtags. As well as strong performance on the hashtag prediction task itself, we show that its learned representation of text (ignoring the hash-tag labels) is useful for other tasks as well. To that end, we present results on a document recommendation task, where it also outperforms a number of baselines.
%@ 978-1-937284-96-1
@inproceedings{conf/emnlp/WestonCA14,
abstract = {We describe a convolutional neural network that learns feature representations for short textual posts using hashtags as a supervised signal. The proposed approach is trained on up to 5.5 billion words predicting 100,000 possible hashtags. As well as strong performance on the hashtag prediction task itself, we show that its learned representation of text (ignoring the hash-tag labels) is useful for other tasks as well. To that end, we present results on a document recommendation task, where it also outperforms a number of baselines.},
added-at = {2017-01-31T11:16:55.000+0100},
author = {Weston, Jason and Chopra, Sumit and Adams, Keith},
biburl = {https://www.bibsonomy.org/bibtex/2ae27449749ffbfddbdcff09b0aa1fa38/nosebrain},
booktitle = {EMNLP},
crossref = {conf/emnlp/2014},
editor = {Moschitti, Alessandro and Pang, Bo and Daelemans, Walter},
ee = {http://aclweb.org/anthology/D/D14/D14-1194.pdf},
interhash = {0ed4314916f8e7c90d066db45c293462},
intrahash = {ae27449749ffbfddbdcff09b0aa1fa38},
isbn = {978-1-937284-96-1},
keywords = {cnn embedding hashtag mining semantic tagspace text},
pages = {1822-1827},
publisher = {ACL},
timestamp = {2017-02-26T17:43:31.000+0100},
title = {#TagSpace: Semantic Embeddings from Hashtags.},
url = {http://dblp.uni-trier.de/db/conf/emnlp/emnlp2014.html#WestonCA14},
year = 2014
}