TweetNLP

A website for developers and researchers to get the most of social media using cutting-edge natural language processing.

Follow us to get TweetNLP updates!🚀

[Jose Camacho Collados @CamachoCollados]

TweetNLP. This is definitely our most popular collection 🔥 With millions of downloads every month and over 500 million downloads overall, this collection contains specialised NLP models for sentiment analysis, emotion detection, offensive language identification and more.

[Jose Camacho Collados @CamachoCollados]

@chesscom Checks out in social media popularity as well! ✅

[Dimos @dimos_antypas]

3/ We evaluated both small PLMs and LLMs using the X-Topic dataset, and found that small, specialised, fine-tuned models can be highly competitive in understanding diverse content. 🚀 Check out the best-performing model here: https://t.co/6Gq47RlG9c

[TweetNLP @tweet_nlp]

Sentiment analysis and much more, up your social media analysis game with the TweetNLP collection!

https://t.co/EcflnHhVUt

[Jose Camacho Collados @CamachoCollados]

Mark Zuckerberg and Yann LeCun boasting about Meta's Llama models being downloaded over 20 million times last month... 😱🤭

[TweetNLP @tweet_nlp]

It's great to see our TweetNLP tools used for this kind of application! 😊

[Uplifting News @UpliftingWeek]

✨ Top 10 positive news in the UK! ✨

🗞️ To start the week with positive energy, we ranked all news stories from last week among the top newspapers in the UK 😀

Here are the winners! 🧵

[TweetNLP @tweet_nlp]

Check out our TweetNLP model collection in Hugging Face! 🤗

👉Sentiment analysis, hate speech detection, topic classification and more!

https://t.co/EcflnHhVUt

[Dimos @dimos_antypas]

🎉 Introducing the SuperTweetEval Collection on Hugging Face! 🤗 Improve your Tweet analysis game with curated datasets and state-of-the-art models. 🚀 Explore all the available resources here 👉 https://t.co/spHYGWsdVP #SuperTweetEval #HuggingFace #NLP

[Jiaxin Pei @jiaxin_pei]

Smartphones are all you need for a poster presentation #EMNLP2023

[TweetNLP @tweet_nlp]

Yesterday at @PyDataBristol, @asahiushio gave a talk about the TweetNLP library! #NLProc #TweetNLP #PyData

[Dimos @dimos_antypas]

🧵 Introducing "SuperTweetEval: A Challenging, Unified and Heterogeneous Benchmark for Social Media NLP Research"

[TweetNLP @tweet_nlp]

Stay tuned as we have updated models for new social media tasks coming very soon! 🔜

https://t.co/DuoWEcAypD

[TweetNLP @tweet_nlp]

Four of the NLP models integrated into TweetNLP have been downloaded over 1 million times in Hugging Face just last month! 🔥

And one of the sentiment analysis models has 10 million monthly downloads!🤯

[TweetNLP @tweet_nlp]

🔥New large-size Twitter LMs available! 🔥

English: https://t.co/pF8r6LAToz
Multilingual: https://t.co/FGLPus2Y3g

🗓️Both trained on data until the end of 2022.

[James Thorne @j6mes]

Pleasure to host @CamachoCollados from @Cardiff_NLP to @kaist_ai to discuss recent work on social media NLP, #tweeteval and #tweetnlp

[TweetNLP @tweet_nlp]

@francksa Sorry we missed this message and thank you for letting us know!

Yes, the platform was temporary unavailable at some point yesterday, but this was already fixed - apologies for the inconvenience!

[TweetNLP @tweet_nlp]

We are excited to officially announce 🚀Tweet Insights🚀, an online visualization platform to get temporal information from Twitter! 📈

Check it out at https://t.co/3JYCM8n2dE

[TweetNLP @tweet_nlp]

@danielbloureiro @Cardiff_NLP @Snap @espinosaanke @lnevespapers @fvancesco @CamachoCollados Bonus! As a byproduct of this project, we release two Twitter-based language models trained until the end of 2022:

🤗RoBERTa-base: https://t.co/sllnlYYAlj

🤗RoBERTa-large: https://t.co/pF8r6LBre7

[TweetNLP @tweet_nlp]

This work has been led by @danielbloureiro, and comes from the always fruitful collaboration between @Cardiff_NLP and @Snap!

There is now an accompanying demo paper summarising the main functionalities: https://t.co/YHO1qdA3Nh

[Dan Lou @danielbloureiro]

If you ever wished there was Google N-grams for Twitter, we have you covered :) It's also much more than frequency stats 📈, including sentiment 🎭, meaning shift 🎇, and diachronic embeddings 🧭 over monthly intervals. Glad to finally share our paper for this 😁 #NLProc