3
$\begingroup$

I am looking to extract the central theme of news headline using NLP/ Text-mining, any references in this direction is of great help.

For example: Inputs:

BRIEF-Dynasil Corporation Of America Reports Q2 EPS Of $0.08

China's night-owl retail investors leverage up to dominate oil futures trade

outputs:

Reports

oil futures

$\endgroup$
0
$\begingroup$

You could formulate the problem as a topic classification task, hence you need labeled data.

From an unsupervised point of view, you could represent sentences with some fixed feature vector (latent representation).

  1. Generating Sentences from a Continuous Space.
  2. Paragraph2Vec.

BRIEF-Dynasil Corporation Of America Reports Q2 EPS Of $0.08

China's night-owl retail investors leverage up to dominate oil futures trade

Self-attention models would be very useful to this kind of problems since you don't need to encode all the context in the last hidden cell of some RNN model to classify to which theme the headline belongs.

| improve this answer | |
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.