3 ms·
Cool idea and execution! Where are you getting the raw article data? Are you doing your own scraping or is it from a 3rd party provider? The couple of tests I
by bvm 11y ago
Cool idea and execution!
Where are you getting the raw article data? Are you doing your own scraping or is it from a 3rd party provider?
The couple of tests I did worked nicely, but it did give me stories that were a few months old, could we have a "don't show me anything older than 24 hours" option?
And finally...any description of what the ML is actually doing - don't want source code or anything, just interested.
- fxgr 11y agoThanks for your feedback! We monitor a pre-defined list of RSS feeds and then send the articles to a document parser that we’ve created. We’re actually thinking of releasing the document parser as an API if there’s interest. A time feature is definitely something we want to add and your suggestion is a cool solution. In terms of the ML my co-founder Ben wrote a non-technical blog post about our approach: https://blog.lateral.io/2015/06/how-do-machines-learn-meaning/ https://blog.lateral.io/2015/06/how-do-machines-learn-meanin...