Read, Attend and Comment: A Deep Architecture for Automatic News Comment Generation by Ze Yang, Can Xu, Wei Wu, Zhoujun Li.
Abstract: Automatic news comment generation is beneficial for real applications but has not attracted enough attention from the research community. In this paper, we propose a “read-attend-comment” procedure for news comment generation and formalize the procedure with a reading network and a generation network. The reading network comprehends a news article and distills some important points from it, then the generation network creates a comment by attending to the extracted discrete points and the news title. We optimize the model in an end-to-end manner by maximizing a variational lower bound of the true objective using the back-propagation algorithm. Experimental results on two public datasets indicate that our model can significantly outperform existing methods in terms of both automatic evaluation and human judgment.
A tweet said this was a “dangerous” paper, so I had to follow the link.
This research could be abused, but how many news comments have you read lately? The comments made by this approach would have to degrade a lot to approach the average human comment.
Anyone who is interested in abusive and/or inane comments, can scrape comments on Facebook or Twitter, set up a cron file and pop off the next comment for posting. Several orders of magnitude less effort that the approach of this paper.
Wondering, would coherence of comments over a large number of articles be an indicator that a bot is involved?