search

yangheng95 / ABSADatasets

Public & Community-shared datasets for Aspect-based sentiment analysis and Text Classification
MIT License
192 stars 64 forks source link

how to handle aspect categories which may not be in the sentence? #32

Open harrywang opened 1 year ago

harrywang commented 1 year ago

Thanks a lot for developing such a nice framework and related tools and datasets.

I have a question about annotation - the aspect terms are from the reviews and some aspect categories might be pre-defined and not in the text.

For example, "the waiter is nice and we spent lots of money on this meal': the waiter and money can be annotated using your tool as aspect terms but what is there are two pre-defined aspect categories such as service and price? Can PyABSA handle this? If yes, how?

Thanks again!

yangheng95 commented 1 year ago

Not available for this now. This case belongs to asepct category detection.

yangheng95 commented 1 year ago

If it is possible, you can contribute the initial implemention of asecpt category detection and datasets in the future.

harrywang commented 1 year ago

If it is possible, you can contribute the initial implemention of asecpt category detection and datasets in the future.

We will try to see what we can do. Thanks!

harrywang commented 1 year ago

Do you by any chance know any Chinese aspect category detection and polarity classification open system/code that we can use? Thanks a lot!

yangheng95 commented 1 year ago

You need to search by yourself on github or elsewhere, there may be no code as easy as this repo.