Much like search engines combat black hat SEO, amazon and reviews systems can do the same.
It would not be to hard too for amazon. Use reviews that are a 'Verified Purchase' and highly 'helpful' reviews as a training/test set. Then machine learn a weighting to every review based on the (product, reviewer, other reviews, etc features...)
The outcome could actually hinder tactics like this for products that game the system much like how link farms hurt sites they were propping up when Google figured out how to stop sites from gaming a search engine.
True, it comparisons do a lot better with more features.
This paper looks to just show the major winning aspect of using CovNets as they do not need many features as the deep net learns its own representations of the training data. It more to show CovNets work on more then just vision.
But architeching the pooling layers IS adding complex to the simple input feature set. Therefore the comparison should be of only state of the art ML.
Any name and shame sites do not last due to the overload of fake reviews and opinions.
The only way is to verify that they stayed there, like what airbnb does or other hoteling and hosteling site do.
This is true of all partner+ or close to partner level management due to that they are there for the long haul. They do not have metrics only ideology.
In the long haul, brand collapse will be seen; see MS where it has trickled from the bottom to the top, hence the exiting CEO and exec shedding.
But who is implementing and setting all the short term 'features', lower mgmt and ICs. The issue is that ideology is not enough for measuring success in mega corps, data has to used. The translation form ideology to metrics is where they fail.
I have no idea why people talk about huge corporations like they have hive mind and are in total unity?
Google is now a mega-corp, comprised of thousands competing for finite success outcomes (raises & promos). It is also data driven hence they look at metrics that are considered success like registers over abandonments. Brand tarnish is a long term outcome, therefore it can not be seen in typical a/b testing cycles. Hence it is ignored, or even gamed upon for short term metric gain.
This is true for all corps that equate success with short cycle data driven metrics.
1. Google is not transparent with its ranking algo. We would need to be.
2. It actually is very cleaning divided, bots & humans. Bad humans have no real power due to point 3.
3. Google has a click traffic amount that is so high that it can only be taken advantage of by bots. Hackernews is tiny in comparison and any algo can be manipulated with just a few bad actors.
Totally agree with you.
It's a game, FB used hard tactics, now he is using public sentiment. It is just an money game in the end. Blog posts like this just hurt both companies perceptions.