TweetNLP

TweetNLP Demo

Sentence/Tweet Classification Hashtag Analysis Word Prediction Sentence/Tweet similarity Named Entity Recognition Question Answering/Generation

Type a sentence or a tweet to get insights (tweet URLs are also accepted)

English, Amharic, Arabic, Armenian, Basque, Bengali, Bulgarian, Burmese, Catalan, Chinese, Czech, Danish, Dhivehi, Dutch, Estonian, Finnish, French, Georgian, German, Greek, Gujarati, Haitian, Hebrew, Hindi, Hungarian, Icelandic, Indonesian, Italian, Japanese, Kannada, Khmer, Korean, Kurdish, Lao, Latvian, Lithuanian, Malayalam, Marathi, Nepali, Norwegian, Oriya, Panjabi, Persian, Polish, Pushto, Romanian, Russian, Serbian, Sindhi, Sinhala, Slovenian, Spanish, Swedish, Tagalog, Tamil, Telugu, Thai, Turkish, Uighur , Ukrainian, Urdu, Vietnamese, Welsh

For Example:

Today is a lovely day!

I really don't like eating vegetables

https://twitter.com/Cardiff_NLP/status/1485518987807137792

Note: This hate speech model is specialised in detecting hate speech towards women and immigrants only.

Note: The stance detection model is currently a proof of concept to assess the stance of the user’s message with respect to abortion.

Note: Tweets get classified into one or more of 19 topics.

Note: This demo predicts the most likely emoji at the end of a given tweet/sentence.

Type a keyword or a hashtag to get an aggregated analysis of what is going on in Twitter these days.

Note: This demo is restricted to the last seven days and ten tweets per day only. For extending this analysis on your own, please check our Tutorial section.

For Example:

#NLProc

Bitcoin

Wales

Start Date:

End Date:

Model:

Language:

Retrieved Tweets ()

Type a sentence or a tweet with a masked word (<mask>) to predict the most likely word. There are models available every three months to select from.

For Example:

I keep forgetting to bring a <mask>.

COVID is a <mask>.

So glad I’m <mask> vaccinated

Looking forward to watching <mask> Game tonight!

Type two sentences (or tweet links) to get their similarity.

Note: The tweet similarity demo is currently available for English only.

For Example:

The elections were won by the liberal candidate.

The weather is going to get cold.

https://twitter.com/CamachoCollados/status/1521827228853673985

For Example:

It's gonna be sunny today.

It's gonna be chilly today.

https://twitter.com/Cardiff_NLP/status/1521825993752813569

Type a sentence or tweet link to get named entities.

The NER demo is currently available for English only.

For Example:

My name is Wolfgang and I live in Berlin

Paradise is a song by Coldplay

https://twitter.com/BBCWorld/status/1532399905217597440

Type a text and a relevant question to get the answer, you can choose Question Generation option and write only your text to get the generated questions and answers.

For Example:

Beyonce further expanded her acting career, starring as blues singer Etta James in the 2008 musical biopic, Cadillac Records.

For Example:

what did beyonce do in the musical biopic?

Output

Type a sentence or a tweet to get insights (tweet URLs are also accepted)

Predictions are based on English (all tasks) or a multilingual model (sentiment). Languages supported are:

Type a keyword or a hashtag to get an aggregated analysis of what is going on in Twitter these days.

Type a sentence or a tweet with a masked word (<mask>) to predict the most likely word. There are models available every three months to select from.

Type two sentences (or tweet links) to get their similarity.

Type a sentence or tweet link to get named entities.

Type a text and a relevant question to get the answer, you can choose Question Generation option and write only your text to get the generated questions and answers.

Sorry!