Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

How useful was the Netflix Prize challenge for Netflix?

quora.com

Post reply on HN
Print view View original

181–190 of 205 posts

Re: How useful was the Netflix Prize challenge for Netflix?

#181

Post by dylan604 »

hateful wrote: post #178
sonofaragorn wrote: post #177

Earlier quoted context omitted.

All these rules determine what it shouldn't recommend. What rules would you use to determine what it should recommend? Would you agree that these rules will quickly become unwieldy and a pain to maintain? And that these will be personalized to your taste but not to someone else's? Wouldn't it be great if you didn't have to maintain those rules and if the system were tailored to every user? Congratulations, you've jus…

Based on the previous posts rules, for me it would be: Recommend anything else you have that's not excluded by those rules. And give me a "remove" like Amazon does and I'll be happy. I don't actually want recommendations, I want a catalog with exclusions.

Eventually, you will be presented a blank page with a bit of text "You have consumed all of the internet. There is no more for you." You can read into that however dark you want to take it.

Re: How useful was the Netflix Prize challenge for Netflix?

#182

Post by capitainenemo »

jmilloy wrote: post #169
mhb wrote: post #167

Earlier quoted context omitted.

Your reference is whatever crap Netflix currently uses. The algorithm they were using at the time of the prize was actually good to begin with. I think they may also have a smaller catalog of movies now. So that wouldn't help.

Yes, the streaming movie catalogue today is probably <1% of the DVD catalogue from the AI challenge days.

The DVD catalogue is still available. They even have an app for impulse adds :) https://dvd.netflix.com/App

Re: How useful was the Netflix Prize challenge for Netflix?

#183

Post by okprod »

koonsolo wrote: post #92

I don't get why they wanted to go all out on AI for this. I use Netflix, and a simple algorithm would already be an improvement: - If I watched a movie and gave it a thumbs down, don't recommend it - If I watched a movie - If I browsed over a movie 50 times, read the info, and still didn't play it, stop recommending it. - If I watched the last episode, remove the "new episodes" banner. WTF

I think we all have personal preferences; for example:

If I watched a movie and gave it a thumbs down, don't recommend it

I sometimes click thumbs down by accident, or upon rewatching change my opinion.

If I watched a movie

I like to rewatch movies, sometimes more than once in a month.

If I browsed over a movie 50 times, read the info, and still didn't play it, stop recommending it.

I have a terrible habit of browsing Netflix and watching trailers right before falling asleep, and I'm sure I've done this to the same titles over and over again.

If I watched the last episode, remove the "new episodes" banner.

I'm not in front of Netflix right now; what happens if new episodes are added while you're watching the series?

Re: How useful was the Netflix Prize challenge for Netflix?

#184

Post by colinmhayes »

oezi wrote: post #154
klausjensen wrote: post #97

Earlier quoted context omitted.

Completely agree. I am constantly baffled at the bullshit Netflix suggests me. I just went to the frontpage of my Netflix and...: - "My list" recommends 3 series where I have alread watched the last episode - "Only on Netflix" recommends another two series, I have already wathed. - A section displayed is "Watch together for older kids" - I dont have kids, never watched any kids stuff on my account. - "Documentaries"…

Related: Has anyone figured out how they determine the trending section? It seems some shows have been trending for years.

Probably a mix of what people are watching and what netflix wants you to watch. The netflix originals are undoubtedly pushed up the trending list.

Re: How useful was the Netflix Prize challenge for Netflix?

#185

Post by PaulHoule »

jareklupinski wrote: post #174
PaulHoule wrote: post #171

Earlier quoted context omitted.

The current "Netflix Optimization Problem" is how to spend the minimum amount on content but still keep you subscribing. The Netflix prize also put Netflix on the map in terms of being a company that solves hard problems. We are still talking about it today and you'd better believe it has inspired talented people to work at Netflix; they could easily blow $1M on recruiting people and have much less to show for it. It…

> The new frontier is that they use your engagement data not just to "suggest" the next movie but to design movies that will keep you engaged. Personally, the opposite has been happening in my recommendations. I've been switching over to Netflix less and less as its library thins, and the replacements seem not as compelling... can't even remember the last time I watched a full movie or TV show season on Netflix. Gues…

"Appealing to narcissism" is dead easy on one level (avoids all sorts of problems that you could encounter with people otherwise), and very hard on another level.

If you are always "present" and engaged then the target is going to do most of the work themselves. If "the lights are on and nobody is home" when you try to take a step forward, you really take 10 steps back.

Re: How useful was the Netflix Prize challenge for Netflix?

#186

Post by BugsJustFindMe »

okprod wrote: post #183
koonsolo wrote: post #92

I don't get why they wanted to go all out on AI for this. I use Netflix, and a simple algorithm would already be an improvement: - If I watched a movie and gave it a thumbs down, don't recommend it - If I watched a movie - If I browsed over a movie 50 times, read the info, and still didn't play it, stop recommending it. - If I watched the last episode, remove the "new episodes" banner. WTF

I think we all have personal preferences; for example: If I watched a movie and gave it a thumbs down, don't recommend it I sometimes click thumbs down by accident, or upon rewatching change my opinion. If I watched a movie I like to rewatch movies, sometimes more than once in a month. If I browsed over a movie 50 times, read the info, and still didn't play it, stop recommending it. I have a terrible habit of browsin…

The engine would be better not doing silly things just because of a "sometimes" fetish or bad faith.

It's also trivial to put all of the things you watched recently into their own subcategory in case you want to watch them again, which is in fact something that Netflix already does. It's called "Watch It Again". There's no reason to pollute recommendations for that.

> I sometimes click thumbs down by accident

The recommendation engine should be obeying your explicit actions, not trying to subvert them. Accidentally clicking thumbs down is an outlier action that is trivial for you to rectify on your own as soon as it happens.

> or upon rewatching change my opinion.

Intentionally rewatching a movie that you expressly disliked is an outlier position.

> I like to rewatch movies, sometimes more than once in a month.

Netflix already has a personal queue+favorites list called "My List" that you can add things to. If it has been less than a month since you last watched something, the reason you're watching it again so soon is because it's on your mind already and you don't need the recommendation engine for that.

Re: How useful was the Netflix Prize challenge for Netflix?

#187

Post by BugsJustFindMe »

bryanrasmussen wrote: post #121
koonsolo wrote: post #92

I don't get why they wanted to go all out on AI for this. I use Netflix, and a simple algorithm would already be an improvement: - If I watched a movie and gave it a thumbs down, don't recommend it - If I watched a movie - If I browsed over a movie 50 times, read the info, and still didn't play it, stop recommending it. - If I watched the last episode, remove the "new episodes" banner. WTF

>- If I watched a movie yeah some movies might be more likely to be rewatchable quicker by a larger amount of the population, and some people might be more likely to be able to rewatch movies they like more quickly than a month - and those people might have preferences that indicate their liking to rewatch more often. Thus it might be nice to do AI on this.

Netflix already gives you a way to add things to a personal list of things you want to watch soon and already has a category for things that you've already seen and enjoyed in case you want to watch them again. There's no need to pollute the general recommendations to scratch this itch.

Re: How useful was the Netflix Prize challenge for Netflix?

#188

Post by jmilloy »

dylan604 wrote: post #180
jmilloy wrote: post #168

Earlier quoted context omitted.

The Netflix catalogue, and the movie rental/streaming ecosystem in general, was completely different when the AI challenge was going on. It was a DVD postal service. Your ratings mattered, and almost any movie you wanted was available. No one was bored scrolling Netflix trying to decide how to kill two hours before bed. They are clearly not using the ratings based automatic recommendations any more. It's not even rel…

>They are clearly not using the ratings based automatic recommendations any more. It's not even relevant given the limited and generally low-quality content available. It's just about keeping enough people paying. For those of us old enough to remember, it's not much different than going to Blockbuster. All of the new releases were along the walls with lots of copies to support the high demand. That's where everyone…

The difference now is that, for 2 or 3 dollars, you can individually rent and stream most movies instantly. No need to pay a subscription that requires you to scroll through a small low-quality subset on a irritating interface.

Re: How useful was the Netflix Prize challenge for Netflix?

#189

Post by jmilloy »

capitainenemo wrote: post #182
jmilloy wrote: post #169

Earlier quoted context omitted.

Yes, the streaming movie catalogue today is probably <1% of the DVD catalogue from the AI challenge days.

The DVD catalogue is still available. They even have an app for impulse adds :) https://dvd.netflix.com/App

Yeah. I finally gave up on it last year not because of the catalogue, but because the turnaround got too slow. Individually renting movies to stream is better all-around.

Re: How useful was the Netflix Prize challenge for Netflix?

#190

Post by sib »

jmilloy wrote: post #188
dylan604 wrote: post #180

Earlier quoted context omitted.

>They are clearly not using the ratings based automatic recommendations any more. It's not even relevant given the limited and generally low-quality content available. It's just about keeping enough people paying. For those of us old enough to remember, it's not much different than going to Blockbuster. All of the new releases were along the walls with lots of copies to support the high demand. That's where everyone…

The difference now is that, for 2 or 3 dollars, you can individually rent and stream most movies instantly. No need to pay a subscription that requires you to scroll through a small low-quality subset on a irritating interface.

Speaking specifically of the US market (I don't know where you are), the list of available titles for transactional rental at any time is a tiny subset of all movies that exist on digital due to windowing (licensing) restrictions. By far, most movies are not available for rental.
Post reply on HN