These are just some thoughts or comments on the non-math parts of the lecture. They are not notes!
Would I pick 2 reviews or 1000? I might if someone whose review I knew and respected was one of the two.
I wonder if Rotten Tomatoes is the same as Amazon? Or Yelp? I actually don't trust Yelp reviews although I do sometimes consult them. Yelp definitely has a gatekeeper problem. The YouTube comments are also completely useless.
Not sure I agree that electronics reviews are objective unless there's a match to specific requirements and objective test measures (no hotter than 98.5, battery lasts 10 hours).
Independence is interesting. On VRBO I felt social pressure to not say anything bad as almost all of the reviews are positive. People are on vacation and only want to say something positive about their fun time.
The look over time charts seem like stock market charts.
It would be helpful in answering the questions if we could easily page back to the lecture. Or is that the point?
Of course, Reddit got IMDb to give top billing to an extra in a movie with Tom Cruise.
3 comments:
Hello! to me, the objective properties of an item are not the only reason to decide to purchase it, may I have a lot of silly or irrational reasons also (just think in Apple fans), so recommendations just give me a "ground" to decide (by example I probably don´t buy something with a lot of bad reports, failure, etc.). Anyway, probably rankings are effective with less selective people or those who are looking sor something to gift
Agreed. At a certain point, a Smartphone is a Smartphone. I'll bet the optimization discussion might have some math to talk about the utility of improvements to rankings. At some point, the rankings are good enough.
Sure!, if Amazon spent such amounts of money of money on it they must be useful. And the irrational impulse to buy may be included at certain degree when reviewers assign the stars.
Anyway Lumix DMC-GX will remain as the camera of my dreams despite is not high in the ranking (sigh)
Post a Comment