Tweet2Vec: Learning Tweet Embeddings Using Character-level CNN-LSTM Encoder-Decoder
We present Tweet2Vec, a novel method for generating general- purpose vector representation of tweets. The model learns tweet embeddings using character-level CNN-LSTM encoder-decoder. We trained our model on 3 million, randomly selected English-language tweets. The model was evaluated using two meth...
| Main Authors: | , , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Language: | en_US |
| Published: |
Association for Computing Machinery (ACM)
2016
|
| Online Access: | http://hdl.handle.net/1721.1/104352 https://orcid.org/0000-0002-2564-8909 https://orcid.org/0000-0002-5826-1591 https://orcid.org/0000-0002-4333-7194 |
| _version_ | 1826204092848930816 |
|---|---|
| author | Vosoughi, Soroush Vijayaraghavan, Prashanth Roy, Deb K |
| author2 | Massachusetts Institute of Technology. Media Laboratory |
| author_facet | Massachusetts Institute of Technology. Media Laboratory Vosoughi, Soroush Vijayaraghavan, Prashanth Roy, Deb K |
| author_sort | Vosoughi, Soroush |
| collection | MIT |
| description | We present Tweet2Vec, a novel method for generating general- purpose vector representation of tweets. The model learns tweet embeddings using character-level CNN-LSTM encoder-decoder. We trained our model on 3 million, randomly selected English-language tweets. The model was evaluated using two methods: tweet semantic similarity and tweet sentiment categorization, outperforming the previous state-of-the-art in both tasks. The evaluations demonstrate the power of the tweet embeddings generated by our model for various tweet categorization tasks. The vector representations generated by our model are generic, and hence can be applied to a variety of tasks. Though the model presented in this paper is trained on English-language tweets, the method presented can be used to learn tweet embeddings for different languages. |
| first_indexed | 2024-09-23T12:48:47Z |
| format | Article |
| id | mit-1721.1/104352 |
| institution | Massachusetts Institute of Technology |
| language | en_US |
| last_indexed | 2024-09-23T12:48:47Z |
| publishDate | 2016 |
| publisher | Association for Computing Machinery (ACM) |
| record_format | dspace |
| spelling | mit-1721.1/1043522022-10-01T11:15:07Z Tweet2Vec: Learning Tweet Embeddings Using Character-level CNN-LSTM Encoder-Decoder Vosoughi, Soroush Vijayaraghavan, Prashanth Roy, Deb K Massachusetts Institute of Technology. Media Laboratory Program in Media Arts and Sciences (Massachusetts Institute of Technology) Vosoughi, Soroush Vosoughi, Soroush Vijayaraghavan, Prashanth Roy, Deb K We present Tweet2Vec, a novel method for generating general- purpose vector representation of tweets. The model learns tweet embeddings using character-level CNN-LSTM encoder-decoder. We trained our model on 3 million, randomly selected English-language tweets. The model was evaluated using two methods: tweet semantic similarity and tweet sentiment categorization, outperforming the previous state-of-the-art in both tasks. The evaluations demonstrate the power of the tweet embeddings generated by our model for various tweet categorization tasks. The vector representations generated by our model are generic, and hence can be applied to a variety of tasks. Though the model presented in this paper is trained on English-language tweets, the method presented can be used to learn tweet embeddings for different languages. 2016-09-20T14:25:48Z 2016-09-20T14:25:48Z 2016-07 Article http://purl.org/eprint/type/ConferencePaper 9781450340694 http://hdl.handle.net/1721.1/104352 Vosoughi, Soroush, Prashanth Vijayaraghavan, and Deb Roy. "Tweet2Vec: Learning Tweet Embeddings Using Character-level CNN-LSTM Encoder-Decoder." Proceedings of the 39th International ACM SIGIR Conference on Research and Development in Information Retrieval - SIGIR ’16, July 17-21, 2016, Pisa, Italy. https://orcid.org/0000-0002-2564-8909 https://orcid.org/0000-0002-5826-1591 https://orcid.org/0000-0002-4333-7194 en_US http://dx.doi.org/10.1145/2911451.2914762 Proceedings of the 39th International ACM SIGIR Conference on Research and Development in Information Retrieval - SIGIR '16 Creative Commons Attribution-Noncommercial-Share Alike http://creativecommons.org/licenses/by-nc-sa/4.0/ application/pdf Association for Computing Machinery (ACM) Vosoughi |
| spellingShingle | Vosoughi, Soroush Vijayaraghavan, Prashanth Roy, Deb K Tweet2Vec: Learning Tweet Embeddings Using Character-level CNN-LSTM Encoder-Decoder |
| title | Tweet2Vec: Learning Tweet Embeddings Using Character-level CNN-LSTM Encoder-Decoder |
| title_full | Tweet2Vec: Learning Tweet Embeddings Using Character-level CNN-LSTM Encoder-Decoder |
| title_fullStr | Tweet2Vec: Learning Tweet Embeddings Using Character-level CNN-LSTM Encoder-Decoder |
| title_full_unstemmed | Tweet2Vec: Learning Tweet Embeddings Using Character-level CNN-LSTM Encoder-Decoder |
| title_short | Tweet2Vec: Learning Tweet Embeddings Using Character-level CNN-LSTM Encoder-Decoder |
| title_sort | tweet2vec learning tweet embeddings using character level cnn lstm encoder decoder |
| url | http://hdl.handle.net/1721.1/104352 https://orcid.org/0000-0002-2564-8909 https://orcid.org/0000-0002-5826-1591 https://orcid.org/0000-0002-4333-7194 |
| work_keys_str_mv | AT vosoughisoroush tweet2veclearningtweetembeddingsusingcharacterlevelcnnlstmencoderdecoder AT vijayaraghavanprashanth tweet2veclearningtweetembeddingsusingcharacterlevelcnnlstmencoderdecoder AT roydebk tweet2veclearningtweetembeddingsusingcharacterlevelcnnlstmencoderdecoder |