How Not to Sort by Average Rating (2009)

Re:How Not to Sort by Average Rating (2009) by ignawin on Wednesday August 30, 2017 @11:23AM (#15132895)

Any blog posts/papers on what the best general approach to onliene reviews is?

[ Parent | on HN ]

Re:How Not to Sort by Average Rating (2009) by dredmorbius on Wednesday August 30, 2017 @01:53PM (#15134344)

Good, qualified, honest reviewers.

Hal Varian (UC Berkeley) has some 1990s refs, which remain good. "Grouplens" is the project/product.

Randy Farmer literally wrote the book on the topic. There's a book, blog, and wiki.

Frankly, Farmer's work, good as it is, largely reinforces my view that Varian captured the essence of the problem, which I've summarised in my opening 'graph. You cannot algorithmically correct for crap quality assessment.

If you're interested in the long-form answer, the fields are epistemology (philosophy) and epistemics (science).

Enjoy!

http://people.ischool.berkeley.edu/~hal/Papers/publish.html

http://people.ischool.berkeley.edu/~ngood/

http://people.ischool.berkeley.edu/~hal/Papers/japan/

http://buildingreputation.com

[ Parent | on HN ]

Re:How Not to Sort by Average Rating (2009) by jules on Thursday August 31, 2017 @03:13AM (#15138699)

I wrote a blog post about a simpler and statistically grounded method: http://julesjacobs.github.io/2015/08/17/bayesian-scoring-of-...

[ Parent | on HN ]