abstractive summary news source?

[AlJohri]

hi @pasmod, just came across your project and trying to better understand it. high level I understand you're scraping news articles in multiple languages- I'm missing to see how you're obtaining attractive summaries from news articles though? are you using news sources that already have human written summaries?

looking through some of the links in appendonly.aof I didn't see any summaries on those pages.

I'm also very interested in a large scale abstractive summary dataset, similar to the CNN/Daily Mail one

[pasmod]

Hi, the headlines of the news articles are considered as a compact abstract summary. Regards, Pashutan

…

On 9. Aug 2018, at 02:44, Al Johri ***@***.***> wrote: hi @pasmod <https://github.com/pasmod>, just came across your project and trying to better understand it. high level I understand you're scraping news articles in multiple languages- I'm missing to see how you're obtaining attractive summaries from news articles though? are you using news sources that already have human written summaries? looking through some of the links in appendonly.aof I didn't see any summaries on those pages. I'm also very interested in a large scale abstractive summary dataset, similar to the CNN/Daily Mail one — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#1>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AEQN0aWHSG_YI5HvEmAqtXPrFzB2Qo_sks5uO4X-gaJpZM4V03-I>.