[luvcraft]

[SuperWes]

My top 5 are all Cave shooters, which is odd because I've never really played a Cave shooter, and certainly haven't rated any. I think it's because my tastes are generally in line with Firenze.

-Wes
_________________

[IIOIOOIOO]

[IIOIOOIOO]

Honestly, my recommendation would be to look through the rankings from that recent netflix competition and hire some single person from the top 20 teams to give you guidance on how to do this. It's hard (very) to get beyond that first "hey, this looks like it might work" point and to a place where your accuracy will be high enough to retain visitors. If you have no money, offer an ownership stake. All of the people behind rank 1 came up with an awesome approach but got nothing for it.

I'd offer myself up but am bound by multiple restrictive agreements as a result of past technology sales. I only offer this guidance freely because it's generally well known and discussed at public presentations, etc.

P.S. You won't need the image thing forever if you make this work. If you are able to drive sales, especially for back-catalog items, publishers will feed you the content in whatever way you like it so that their stuff will appear on your site.

[IIOIOOIOO]

PPS: Cloud Tags are not meaningful meta-data. They're cool for natural search stuff but terrible for quantitative analysis.

[BenoitRen]

Achievements? *vomits*
_________________
Get Xenoblade Chronicles!

[smartblue]

You feeling okay, Benren? Usually you are spitting out quote boxes by now.
_________________

[luvcraft]

[BenoitRen]

[Endless]

Message bugged out. Basically saying all images should be rectangles, not just top row.

[remote]

Just throwing in a vote saying I think the image sizing is fine as it is. I like seeing the whole cover, nice and big.
_________________

letterboxd | last.fm | steam

[SuperWes]

[luvcraft]

[IIOIOOIOO]

Luvcraft, please: Let me get this one for you.

But seriously, and I mean no offense by this, the likelihood is high that your result accuracy will degrade as your network grows more dense. Both as more users enter the system and as individuals rate more games. Your accuracy will end up lower than 20% r-squared. Eventually, you will end up with everyone being recommended hentai games because they like STGs. You will struggle spectacularly with people who have multiple niche genre interests as those interests largely have poor inferential value from one to the next. As I said, it appears (from reported false hits) that your engine largely relies on "similar profile inference."

If you intend to deliver this as a marketable product, the quality of the engine will need to be much better. Services like Netflix and Amazon have set the bar pretty high in this arena. Consumers have grown used to engines that "just work" and will not be pleased by poor performance in this area. Will your engine ever be able to predict whether a person will enjoy a game that just came out and almost nobody in your network has played? Nope. Will your engine ever be able to understand that a user JUST DOESN'T LIKE some type of game? I'm sure you're a great programmer but this is a difficult problem you're trying to solve and you seem resistant to feedback from a person who actually knows how to do what you're trying to do. Find someone to help you, even if it means diluting your ownership stake. Giving away 25% of all future profits on a service that actually grows is better than keeping 100% of profits from a venture which goes nowhere.

I appreciate new projects and I very much hope that you succeed. Please be willing to learn from those that came this way before you.

[luvcraft]

Sorry, some combination of the way you wrote that first reply and the way I read it make me think you didn't know what you were talking about, and had never actually used reccr. This new reply is a lot more coherent.

[IIOIOOIOO]

Sorry if I was confusing. English is not my first (or third) language.

Here's a good exercise to better understand. Get your hands on the Netflix sample data. Apply your algorithm using no metadata beyond Title and Format. Break the sample set into 5% chunks and feed your engine increasingly large sets (5, 10, 15, etc) while applying the resultant information to the prediction test files included. Up through 25% of set, it probably gets better (because it goes from random to something.) Beyond that, accuracy will likely begin decreasing again.

Associative Set-based prediction engines that only use network selections (overlap between liked games between set participants) more or less always coalesce into highly segregated spires of recommendation overlaps and ultimately fail in their ability to distinguish "dislikes" unless there is a sufficient strong relationship between the core area of overlap and tangential interests. I.E. If all people who like STGs also like H-Games, your engine will do just fine. If not, accuracy will begin to degrade as "Set Megatrends" (Sorry, struggling for a better way to describe w/o maths) emerge through the growth of your network. Your problem is further complicated by not having any metadata to allow for matrix/vector based to distinguish characteristics useful for predicting segmenting factors.

I should have better-clarified my challenge: Can your engine predict with anything above random accuracy that a user will like a game that NOBODY has rated yet? Everyone else's can.

Is there no source for the necessary metadata? If not, and if this is a commercial venture, consider using the Mechanical Turk to gather it. You can probably collect that info for USD .10 per title and also have the opportunity to resell it.

I've invested more time than intended in this conversation. I am not going to solve this problem for you, for various reasons. As a starting point for your exploration, check out this article:
http://blog.smellthedata.com/2009/06/netflix-prize-tribute-recommendation.html

I think it does a good job of taking a complex problem and posing it in simple terms... especially for a developer. Read it, and the commentary, and you will see several techniques named which are worth investigating.

Welcome to the Internet. All of the easy problems were solved in the 90s.

[IIOIOOIOO]

Also, good luck. I look on with interest.

[TheUser]

I just now signed up and haven't fooled around with it much, but I think having a way to export your ratings would be nice. I tend to feel a bit reserved about putting a lot of information into any database that isn't tied to my own site or backed up by me somehow. That's not to say I won't use it, but it's nice to be able save a list of my data.

edit(s):

How are you treating DLC like GTA4's Lost and the Damned and the Ballad of Gay Tony? Those are just examples.

N+ for DS and PSP are very different.

Do you do any notifications when/if submitted games are added? I'd like to know what to come back and rate, but I wasn't keeping track of my submissions.

[luvcraft]

Wow, TheUser, thanks for the 136 new submissions! :)

[luvcraft]

[TheUser]

[luvcraft]

[luvcraft]

OK, I've added the "Featured Games" feature to reccr, which is basically as close as I intend reccr to get to advertising. I want more games to fill out the "Featured Games" roster, so if you're somehow involved in the development or publication of a game and would like to have it featured for free for a couple months, let me know! I'm making a special effort NOT to give the games I've made any preferential treatment on reccr, so I'm keeping my games out of the featured games list until I reach the totally arbitrary number of 10 other featured games.
_________________

[TheUser]

I'm definitely enjoying playing with the system. It's recommending me a bunch of stuff with which I'm unfamiliar - Mushihime-sama Futari, Espgaluda II Black Label, Ninja Bros. 忍者ブラザーズ, Jazz Jackrabbit 2: The Secret Files, and Johnny Platform's Biscuit Romp are my top five. I should try some of the games it recommends to see how good it's doing.

I ended up not sleeping much last night because I was rating games late. I seem to have an obsession with list-keeping and stat tracking. I keep my game information on my own site (http://www.webpageless.net/gaming/games.shtml), which isn't pretty but is made just for myself.

Now, which 136 games were those that I submitted so I can go rate them...

[luvcraft]

[luvcraft]

[luvcraft]

Alright my SB peoples, I've finally caved and added 8 more platforms to reccr, going all the way back to 16-bit systems (except the Jaguar, because LOL). So if there are some psx / snes / genesis games you've been aching to rate, now you can (but you'll probably have to add them first unless someone else already added them).
_________________

[protoblax]

As an experiment, I'm gonna plug in all the scores from actionbutton and see what comes up. I will give 0s and .5s one star, 1s and 1.5s 2 stars, and so on. Only perfect 4s get 5 stars.
_________________

[protoblax]

Experiment complete! Top ten games recommended to ABDN by reccr:

1. Super Castlevania IV
2. Armored Core
3. Espgaluda II Black Label
4. Skies of Arcadia
5. Company of Heroes: Tales of Valor
6. Ketsui: Kizuna Jigoku Tachi EXTRA
7. Final Fantasy VI
8. Startopia
9. DoDonPachi Daioujou Black Label Extra
10. Space Funeral

Hm, not a bad list.
_________________

[luvcraft]

cool!

interesting that there are so many shmups in there.

[protoblax]

Updated, because I missed a few:

1. King's Field
2. Espgaluda II Black Label
3. Company of Heroes: Tales of Valor
4. Ketsui: Kizuna Jigoku Tachi EXTRA
5. Ys I And II Eternal Story
6. Gravity Bone
7. Armored Core
8. Skies of Arcadia
9. Startopia
10. DoDonPachi Daioujou Black Label Extra
_________________

[luvcraft]

Yeah, Gravity Bone just jumped to the top of my list, too. Guess I should check it out!
_________________

[Bennett]

It's interesting. I don't deny that Espgaluda II, Super Castlevania IV and Final Fantasy VI are good games, but as someone who came to all of them quite a long time after their release, largely on SBDN recommendations, they really didn't move me at all. It's not that way for all old games, though — I never played Symphony of the Night until this year, again on an SBDN recommendation, and it blew me away.

Gotta try King's Field, finally.

[luvcraft]

[protoblax]

The list [and war] has changed...

1. Mushihime-sama Futari
2. King's Field
3. Espgaluda II Black Label
4. Company of Heroes: Tales of Valor
5. Ketsui: Kizuna Jigoku Tachi EXTRA
6. Ys I And II Eternal Story
7. Tomba!
8. Armored Core
9. Skies of Arcadia
10. Gravity Bone
_________________

[remote]

My recs are fairly similar.

1. Armored Core: For Answer
2. GTA IV: The Ballad of Gay Tony
3. Company of Heroes: Opposing Fronts
4. Monster Hunter Freedom Unite
5. Gravity Bone
6. Company of Heroes: Tales of Valor
7. Tomba!
8. bit Generations: Soundvoyager
9. Saira
10. Space Funeral

Interesting that The Ballad of Gay Tony is up there, but not The Lost and the Damned. Will try them both soon enough...!

Is that the best Monster Hunter title? I don't have a PSP, but I might like to try one of those some time.
_________________

letterboxd | last.fm | steam

[Sniper Honeyviper]

Portable 3rd is probably the best, simply because it is the newest. I hear they finally made the Felyne companions useful!

A US release is kind of doubtful at this point, though.

[luvcraft]

[remote]

Is there any way to change the twitter account I have connected to my reccr account?
_________________

letterboxd | last.fm | steam

[luvcraft]

[parkbench]

could you link mobygames in each profile? i find it way more substantive and useful than indigames db or the bland amazon link. mobygames is an old standby of games encyclopedia-ing so it may not be a bad idea. HOTU might also be a good augment for some of the more obscure pc games.

seriously, i think improving the resources you build reccr upon could really strengthen it. i'm enjoying it so far as someone who has distanced themselves from being actively involved in gaming communities, who just wants good recommendations now and then. i'm a perfect target user.

i will say that i share some of the discontent around the default assumption that it's good to use social networking sites/other accounts like google to sign in. i know a lot of thought was put into it and that there are reasons behind it, it's just that there is a larger discourse around the supposed advantage of this kind of 'inter-operability', which practically means more datamining and less privacy by default. that might sound like a crude line, but i think we will all see the effects of this in the coming years and wonder how we got there--exactly assumptions about web design like this. right now i think it's basically a harmless but unfortunate decision, but it can always come back around to bite you in the ass in the long run, i think.

edit: just a note it's giving me MAD GOOD recommendations though, kudos to you for that.
_________________
metafilter vs. youtube comments