BGG Wiki - Revision History - Game Recommendation Algorithm

Game Recommendation Algorithm: - removed introduced date (it was sometime in 2005 or 2006) and elaborated

2008-06-22T15:06:11-05:00

From TVis 2008-03-12 08:05:29		To reapersaurus 2008-06-22 15:06:11


	===Previous Game Recommendation system===		===Previous Game Recommendation system===
-	Introduced January 2008, this algorithm replaced the previous recommendation system which was user-entered. People could go to the recommendation section, and add games that they thought people who liked this game would enjoy. The more recommendations a game had, the higher ranked corrolation it would get. Users could even disagree with recommendations, which easily allowed for user corrections.	+	This algorithm replaced the previous recommendation system which was user-entered (pre-2005).
		+	Interested members could go to the recommendation section, and add games that they thought people who liked this game would enjoy. The more recommendations a game had, the higher ranked corrolation it would get. Members could even disagree with recommendations, which easily allowed for user corrections.
		+	It worked for games that had enough interested members to enter recommendations, but didn't work for games with not enough interested members. This was the reason why a computerized algorithm was suggested as a replacement - to provide recommendations for all games (even though it still has a threshold requirement).

Game Recommendation Algorithm: - line break edit

2008-03-12T08:05:29-05:00

From TVis 2008-03-12 08:02:29		To TVis 2008-03-12 08:05:29


	===Previous Game Recommendation system===		===Previous Game Recommendation system===
-	Inroduced January 2008, this algorithm replaced the previous recommendation system, which was user-entered.	+	Introduced January 2008, this algorithm replaced the previous recommendation system which was user-entered. People could go to the recommendation section, and add games that they thought people who liked this game would enjoy. The more recommendations a game had, the higher ranked corrolation it would get. Users could even disagree with recommendations, which easily allowed for user corrections.
-	People could go to the recommendation section, and add games that they thought people who liked this game would enjoy. The more recommendations a game had, the higher ranked corrolation it would get. Users could even disagree with recommendations, which easily allowed for user corrections.	+

Game Recommendation Algorithm: - Introduction date

2008-03-12T08:02:29-05:00

From reapersaurus 2007-12-04 12:00:28		To TVis 2008-03-12 08:02:29


	===Previous Game Recommendation system===		===Previous Game Recommendation system===
-	This algorithm replaced the previous recommendation system, which was user-entered.	+	Inroduced January 2008, this algorithm replaced the previous recommendation system, which was user-entered.
	People could go to the recommendation section, and add games that they thought people who liked this game would enjoy. The more recommendations a game had, the higher ranked corrolation it would get. Users could even disagree with recommendations, which easily allowed for user corrections.		People could go to the recommendation section, and add games that they thought people who liked this game would enjoy. The more recommendations a game had, the higher ranked corrolation it would get. Users could even disagree with recommendations, which easily allowed for user corrections.

Game Recommendation Algorithm: - added historical system information

2007-12-04T12:00:28-06:00

From glsonn 2007-09-24 23:58:23		To reapersaurus 2007-12-04 12:00:28

	===Poorly rated/unowned Games===		===Poorly rated/unowned Games===

-	As is probably obvious from the "how it works" section, if a game is owned by a very small sample of people, or liked by a very small sample of people, it is difficult to provide recommendations. There simply aren't enough other people who like it to come up with a fair generalization of what you will like "if you like Poorly Rated Game". This isn't really much of a problem, in that weak recommendations are still generated, and the effect is only severe under games that almost no one actually likes, so there are few who will say "I like this. What do you recommend?"	+	As is probably obvious from the "how it works" section, if a game is owned by a very small sample of people, or liked by a very small sample of people, it is difficult to provide recommendations. There simply aren't enough other people who like it to come up with a fair generalization of what you will like "if you like Poorly Rated Game". This isn't really much of a problem, in that weak recommendations are still generated, and the effect is only severe under games that almost no one actually likes, so there are few who will say "I like this. What do you recommend?"
		+
		+	===Previous Game Recommendation system===
		+	This algorithm replaced the previous recommendation system, which was user-entered.
		+	People could go to the recommendation section, and add games that they thought people who liked this game would enjoy. The more recommendations a game had, the higher ranked corrolation it would get. Users could even disagree with recommendations, which easily allowed for user corrections.

Game Recommendation Algorithm: - No Comment

2007-09-24T23:58:23-05:00

From Nightmare 2007-01-16 18:15:56		To glsonn 2007-09-24 23:58:23

	==How it works==		==How it works==

-	Take a game, call it GameX. Say that game is owned by 1% of people. Take another game, GameY, which is owned by 300 people. You'd expect about 3 of them to also own GameX. If many more do own GameX, that suggests that people who like GameY enough to own it like GameX enough to own it '''more often than most'''. Now, if only 4 people own it, that's not such an anomoly. If 30 people own it, that's a dramatic factor of 10. In contrast, another game, GameZ, owned by 10 people, you'd expect noone (well, 0.1 people) to own GameY. But if a single person does, that's also a factor of 10, but it isn't as genuine of an anomaly. To correct for this, the algorithm adjusts for the relative obscurity of the game.	+	Take a game, call it GameX. Say that game is owned by 1% of people. Take another game, GameY, which is owned by 300 people. You'd expect about 3 of them to also own GameX. If many more do own GameX, that suggests that people who like GameY enough to own it like GameX enough to own it '''more often than most'''. Now, if only 4 people own it, that's not such an anomaly. If 30 people own it, that's a dramatic factor of 10. In contrast, another game, GameZ, owned by 10 people, you'd expect no one (well, 0.1 people) to own GameY. But if a single person does, that's also a factor of 10, but it isn't as genuine of an anomaly. To correct for this, the algorithm adjusts for the relative obscurity of the game.

	A similar process is applied to game ratings, and these two rates, along with the obscurity correction combine to produce the score.		A similar process is applied to game ratings, and these two rates, along with the obscurity correction combine to produce the score.

	===New Games===		===New Games===

-	New games typically only recommend other new games. This is because people who own new games have often placed a recent order, and have anomolously high ownership/ratership of other new games. Once the game becomes more widely available, this tends to go away.	+	New games typically only recommend other new games. This is because people who own new games have often placed a recent order, and have anomalously high ownership/ratership of other new games. Once the game becomes more widely available, this tends to go away.

	Games which were only available at a particular very limited time or place have the same problem to some degree.		Games which were only available at a particular very limited time or place have the same problem to some degree.

	===Poorly rated/unowned Games===		===Poorly rated/unowned Games===

-	As is probably obvious from the "how it works" section, if a game is owned by a very small sample of people, or liked by a very small sample of people, it is difficult to provide recommendations. There simply aren't enough other people who like it to come up with a fair generalization of what you will like "if you like Poorly Rated Game". This isn't really much of a problem, in that weak recommendations are still generated, and the effect is only severe under games that almost noone actually likes, so there are few who will say "I like this. What do you recommend?"	+	As is probably obvious from the "how it works" section, if a game is owned by a very small sample of people, or liked by a very small sample of people, it is difficult to provide recommendations. There simply aren't enough other people who like it to come up with a fair generalization of what you will like "if you like Poorly Rated Game". This isn't really much of a problem, in that weak recommendations are still generated, and the effect is only severe under games that almost no one actually likes, so there are few who will say "I like this. What do you recommend?"

Game Recommendation Algorithm: - No Comment

2007-01-16T18:15:56-06:00

From wtrollkin2000 2007-01-05 10:29:48		To Nightmare 2007-01-16 18:15:56

	==How it works==		==How it works==

-	Take a game, call it GameX. Say that game is owned by 1% of people. Take another game, GameY, which is owned by 300 people. You'd expect about 3 of them to also own GameX. If many more do own GameX, that suggests that people who like GameY enough to own it like GameX enough to own it '''more often than most'''. Now, if only 4 people own it, that's not such an anomoly. If 30 people own it, that's a dramatic factor of 10. In contrast, another game, GameZ, owned by 10 people, you'd expect noone (well, 0.1 people) to one GameY. But if a single person does, that's also a factor of 10, but it isn't as genuine of an anomaly. To correct for this, the algorithm adjusts for the relative obscurity of the game.	+	Take a game, call it GameX. Say that game is owned by 1% of people. Take another game, GameY, which is owned by 300 people. You'd expect about 3 of them to also own GameX. If many more do own GameX, that suggests that people who like GameY enough to own it like GameX enough to own it '''more often than most'''. Now, if only 4 people own it, that's not such an anomoly. If 30 people own it, that's a dramatic factor of 10. In contrast, another game, GameZ, owned by 10 people, you'd expect noone (well, 0.1 people) to own GameY. But if a single person does, that's also a factor of 10, but it isn't as genuine of an anomaly. To correct for this, the algorithm adjusts for the relative obscurity of the game.

	A similar process is applied to game ratings, and these two rates, along with the obscurity correction combine to produce the score.		A similar process is applied to game ratings, and these two rates, along with the obscurity correction combine to produce the score.

Game Recommendation Algorithm: - elaborate on same year effect

2007-01-05T10:29:48-06:00

From mkgray 2006-12-31 23:10:36		To wtrollkin2000 2007-01-05 10:29:48

	===Same Year/Publisher Effect===		===Same Year/Publisher Effect===

-	Because games that come out in the same year are likely to share owners and raters by virtue of simultaneous availability, the system tends to be biased towards games that came out at the same time. This is to be expected; If you like rail games, you're likely to buy more rail games. Similarly, if you started buying games in 2001, you're likely to buy more games in 2001 than 1991.	+	Because games that come out in the same year are likely to share owners and raters by virtue of simultaneous availability, the system tends to be biased towards games that came out at the same time. This is to be expected; If you like rail games, you're likely to buy more rail games. Similarly, if you started buying games in 2001, you're likely to buy more games in 2001 than 1991. Also one time raters (be they never returning BGG members or not) will create a snapshot of the games that are in vogue during their data entry time frame

	The same issue causes games by the same publisher to be frequently recommended. Of course, publishers often publish similar games to others they've published, to "play to the base", so is even less of an issue.		The same issue causes games by the same publisher to be frequently recommended. Of course, publishers often publish similar games to others they've published, to "play to the base", so is even less of an issue.

Game Recommendation Algorithm: - No Comment

2006-12-31T23:10:36-06:00

From mkgray 2006-12-31 22:54:39		To mkgray 2006-12-31 23:10:36

	===Very Popular Games===		===Very Popular Games===

-	For games that are extremely popular and nearly universally well liked, such as [gameid=3076] (yes, there are those that don't like it; I said "nearly"), the system can't give anything other than very weak recommendations. Saying "If you like Puerto Rico, you'll also like..." is (statistically) very hard to discern, because, well, the first part ("If you like Puerto Rico") includes nearly everyone, therefore there's nothing characteristically different about "everyone".	+	For games that are extremely popular and nearly universally well liked, such as [gameid=3076] (yes, there are those that don't like it; I said "nearly"), the system can't give anything other than very weak recommendations. Saying "If you like Puerto Rico, you'll also like..." is (statistically) very hard to discern, because, well, the first part ("If you like Puerto Rico") includes nearly everyone, therefore there's nothing characteristically different about "everyone".
		+
		+	===Poorly rated/unowned Games===
		+
		+	As is probably obvious from the "how it works" section, if a game is owned by a very small sample of people, or liked by a very small sample of people, it is difficult to provide recommendations. There simply aren't enough other people who like it to come up with a fair generalization of what you will like "if you like Poorly Rated Game". This isn't really much of a problem, in that weak recommendations are still generated, and the effect is only severe under games that almost noone actually likes, so there are few who will say "I like this. What do you recommend?"

Game Recommendation Algorithm: - No Comment

2006-12-31T22:54:39-06:00

From mkgray 2006-12-31 20:54:54		To mkgray 2006-12-31 22:54:39

	===New Games===		===New Games===

-	New games typically only recommend other new games. This is because people who own new games have often placed a recent order, and have anomolously high ownership/ratership of other new games. Once the game becomes more widely available, this tends to go away.	+	New games typically only recommend other new games. This is because people who own new games have often placed a recent order, and have anomolously high ownership/ratership of other new games. Once the game becomes more widely available, this tends to go away.
		+
		+	Games which were only available at a particular very limited time or place have the same problem to some degree.
		+
		+	===Very Popular Games===
		+
		+	For games that are extremely popular and nearly universally well liked, such as [gameid=3076] (yes, there are those that don't like it; I said "nearly"), the system can't give anything other than very weak recommendations. Saying "If you like Puerto Rico, you'll also like..." is (statistically) very hard to discern, because, well, the first part ("If you like Puerto Rico") includes nearly everyone, therefore there's nothing characteristically different about "everyone".

Game Recommendation Algorithm: - No Comment

2006-12-31T20:54:54-06:00

From mkgray 2006-12-31 20:54:34		To mkgray 2006-12-31 20:54:54

	On the [[game entry]] page for each game, a list of recommendations is generated. These recommendations are based on user ratings and user collections. The strength of the recommendation is calculated into a score, and the games that get the highest score are listed on the game page.		On the [[game entry]] page for each game, a list of recommendations is generated. These recommendations are based on user ratings and user collections. The strength of the recommendation is calculated into a score, and the games that get the highest score are listed on the game page.

-	==How it works===	+	==How it works==

	Take a game, call it GameX. Say that game is owned by 1% of people. Take another game, GameY, which is owned by 300 people. You'd expect about 3 of them to also own GameX. If many more do own GameX, that suggests that people who like GameY enough to own it like GameX enough to own it '''more often than most'''. Now, if only 4 people own it, that's not such an anomoly. If 30 people own it, that's a dramatic factor of 10. In contrast, another game, GameZ, owned by 10 people, you'd expect noone (well, 0.1 people) to one GameY. But if a single person does, that's also a factor of 10, but it isn't as genuine of an anomaly. To correct for this, the algorithm adjusts for the relative obscurity of the game.		Take a game, call it GameX. Say that game is owned by 1% of people. Take another game, GameY, which is owned by 300 people. You'd expect about 3 of them to also own GameX. If many more do own GameX, that suggests that people who like GameY enough to own it like GameX enough to own it '''more often than most'''. Now, if only 4 people own it, that's not such an anomoly. If 30 people own it, that's a dramatic factor of 10. In contrast, another game, GameZ, owned by 10 people, you'd expect noone (well, 0.1 people) to one GameY. But if a single person does, that's also a factor of 10, but it isn't as genuine of an anomaly. To correct for this, the algorithm adjusts for the relative obscurity of the game.

Game Recommendation Algorithm: - No Comment

2006-12-31T20:54:34-06:00

From mkgray 2006-12-31 20:54:00		To mkgray 2006-12-31 20:54:34

	On the [[game entry]] page for each game, a list of recommendations is generated. These recommendations are based on user ratings and user collections. The strength of the recommendation is calculated into a score, and the games that get the highest score are listed on the game page.		On the [[game entry]] page for each game, a list of recommendations is generated. These recommendations are based on user ratings and user collections. The strength of the recommendation is calculated into a score, and the games that get the highest score are listed on the game page.

-	===How it works===	+	==How it works===

	Take a game, call it GameX. Say that game is owned by 1% of people. Take another game, GameY, which is owned by 300 people. You'd expect about 3 of them to also own GameX. If many more do own GameX, that suggests that people who like GameY enough to own it like GameX enough to own it '''more often than most'''. Now, if only 4 people own it, that's not such an anomoly. If 30 people own it, that's a dramatic factor of 10. In contrast, another game, GameZ, owned by 10 people, you'd expect noone (well, 0.1 people) to one GameY. But if a single person does, that's also a factor of 10, but it isn't as genuine of an anomaly. To correct for this, the algorithm adjusts for the relative obscurity of the game.		Take a game, call it GameX. Say that game is owned by 1% of people. Take another game, GameY, which is owned by 300 people. You'd expect about 3 of them to also own GameX. If many more do own GameX, that suggests that people who like GameY enough to own it like GameX enough to own it '''more often than most'''. Now, if only 4 people own it, that's not such an anomoly. If 30 people own it, that's a dramatic factor of 10. In contrast, another game, GameZ, owned by 10 people, you'd expect noone (well, 0.1 people) to one GameY. But if a single person does, that's also a factor of 10, but it isn't as genuine of an anomaly. To correct for this, the algorithm adjusts for the relative obscurity of the game.

	A similar process is applied to game ratings, and these two rates, along with the obscurity correction combine to produce the score.		A similar process is applied to game ratings, and these two rates, along with the obscurity correction combine to produce the score.

-	===Known issues===	+	==Known issues==

	The algorithm is far from perfect, but produces pretty good results. It does have some known behaviors which are worth noting.		The algorithm is far from perfect, but produces pretty good results. It does have some known behaviors which are worth noting.

-	==Same Year/Publisher Effect==	+	===Same Year/Publisher Effect===

	Because games that come out in the same year are likely to share owners and raters by virtue of simultaneous availability, the system tends to be biased towards games that came out at the same time. This is to be expected; If you like rail games, you're likely to buy more rail games. Similarly, if you started buying games in 2001, you're likely to buy more games in 2001 than 1991.		Because games that come out in the same year are likely to share owners and raters by virtue of simultaneous availability, the system tends to be biased towards games that came out at the same time. This is to be expected; If you like rail games, you're likely to buy more rail games. Similarly, if you started buying games in 2001, you're likely to buy more games in 2001 than 1991.

	The same issue causes games by the same publisher to be frequently recommended. Of course, publishers often publish similar games to others they've published, to "play to the base", so is even less of an issue.		The same issue causes games by the same publisher to be frequently recommended. Of course, publishers often publish similar games to others they've published, to "play to the base", so is even less of an issue.

-	==Favorites==	+	===Favorites===

	For reasons not yet entirely obvious, the algorithm seems to have some "favorites", or games that get recommended a lot. You'll see Showmanager and Medici, for example, on a lot of lists.		For reasons not yet entirely obvious, the algorithm seems to have some "favorites", or games that get recommended a lot. You'll see Showmanager and Medici, for example, on a lot of lists.

-	==New Games==	+	===New Games===

	New games typically only recommend other new games. This is because people who own new games have often placed a recent order, and have anomolously high ownership/ratership of other new games. Once the game becomes more widely available, this tends to go away.		New games typically only recommend other new games. This is because people who own new games have often placed a recent order, and have anomolously high ownership/ratership of other new games. Once the game becomes more widely available, this tends to go away.

Game Recommendation Algorithm: - initial version

2006-12-31T20:54:00-06:00

From reapersaurus 2008-06-22 15:06:11		To mkgray 2006-12-31 20:54:00

	On the [[game entry]] page for each game, a list of recommendations is generated. These recommendations are based on user ratings and user collections. The strength of the recommendation is calculated into a score, and the games that get the highest score are listed on the game page.		On the [[game entry]] page for each game, a list of recommendations is generated. These recommendations are based on user ratings and user collections. The strength of the recommendation is calculated into a score, and the games that get the highest score are listed on the game page.

-	==How it works==	+	===How it works===

-	Take a game, call it GameX. Say that game is owned by 1% of people. Take another game, GameY, which is owned by 300 people. You'd expect about 3 of them to also own GameX. If many more do own GameX, that suggests that people who like GameY enough to own it like GameX enough to own it '''more often than most'''. Now, if only 4 people own it, that's not such an anomaly. If 30 people own it, that's a dramatic factor of 10. In contrast, another game, GameZ, owned by 10 people, you'd expect no one (well, 0.1 people) to own GameY. But if a single person does, that's also a factor of 10, but it isn't as genuine of an anomaly. To correct for this, the algorithm adjusts for the relative obscurity of the game.	+	Take a game, call it GameX. Say that game is owned by 1% of people. Take another game, GameY, which is owned by 300 people. You'd expect about 3 of them to also own GameX. If many more do own GameX, that suggests that people who like GameY enough to own it like GameX enough to own it '''more often than most'''. Now, if only 4 people own it, that's not such an anomoly. If 30 people own it, that's a dramatic factor of 10. In contrast, another game, GameZ, owned by 10 people, you'd expect noone (well, 0.1 people) to one GameY. But if a single person does, that's also a factor of 10, but it isn't as genuine of an anomaly. To correct for this, the algorithm adjusts for the relative obscurity of the game.

	A similar process is applied to game ratings, and these two rates, along with the obscurity correction combine to produce the score.		A similar process is applied to game ratings, and these two rates, along with the obscurity correction combine to produce the score.

-	==Known issues==	+	===Known issues===

	The algorithm is far from perfect, but produces pretty good results. It does have some known behaviors which are worth noting.		The algorithm is far from perfect, but produces pretty good results. It does have some known behaviors which are worth noting.

-	===Same Year/Publisher Effect===	+	==Same Year/Publisher Effect==

-	Because games that come out in the same year are likely to share owners and raters by virtue of simultaneous availability, the system tends to be biased towards games that came out at the same time. This is to be expected; If you like rail games, you're likely to buy more rail games. Similarly, if you started buying games in 2001, you're likely to buy more games in 2001 than 1991. Also one time raters (be they never returning BGG members or not) will create a snapshot of the games that are in vogue during their data entry time frame	+	Because games that come out in the same year are likely to share owners and raters by virtue of simultaneous availability, the system tends to be biased towards games that came out at the same time. This is to be expected; If you like rail games, you're likely to buy more rail games. Similarly, if you started buying games in 2001, you're likely to buy more games in 2001 than 1991.

	The same issue causes games by the same publisher to be frequently recommended. Of course, publishers often publish similar games to others they've published, to "play to the base", so is even less of an issue.		The same issue causes games by the same publisher to be frequently recommended. Of course, publishers often publish similar games to others they've published, to "play to the base", so is even less of an issue.

-	===Favorites===	+	==Favorites==

	For reasons not yet entirely obvious, the algorithm seems to have some "favorites", or games that get recommended a lot. You'll see Showmanager and Medici, for example, on a lot of lists.		For reasons not yet entirely obvious, the algorithm seems to have some "favorites", or games that get recommended a lot. You'll see Showmanager and Medici, for example, on a lot of lists.

-	===New Games===	+	==New Games==

-	New games typically only recommend other new games. This is because people who own new games have often placed a recent order, and have anomalously high ownership/ratership of other new games. Once the game becomes more widely available, this tends to go away.	+	New games typically only recommend other new games. This is because people who own new games have often placed a recent order, and have anomolously high ownership/ratership of other new games. Once the game becomes more widely available, this tends to go away.
-		+
-	Games which were only available at a particular very limited time or place have the same problem to some degree.	+
-		+
-	===Very Popular Games===	+
-		+
-	For games that are extremely popular and nearly universally well liked, such as [gameid=3076] (yes, there are those that don't like it; I said "nearly"), the system can't give anything other than very weak recommendations. Saying "If you like Puerto Rico, you'll also like..." is (statistically) very hard to discern, because, well, the first part ("If you like Puerto Rico") includes nearly everyone, therefore there's nothing characteristically different about "everyone".	+
-		+
-	===Poorly rated/unowned Games===	+
-		+
-	As is probably obvious from the "how it works" section, if a game is owned by a very small sample of people, or liked by a very small sample of people, it is difficult to provide recommendations. There simply aren't enough other people who like it to come up with a fair generalization of what you will like "if you like Poorly Rated Game". This isn't really much of a problem, in that weak recommendations are still generated, and the effect is only severe under games that almost no one actually likes, so there are few who will say "I like this. What do you recommend?"	+
-		+
-	===Previous Game Recommendation system===	+
-	This algorithm replaced the previous recommendation system which was user-entered (pre-2005).	+
-	Interested members could go to the recommendation section, and add games that they thought people who liked this game would enjoy. The more recommendations a game had, the higher ranked corrolation it would get. Members could even disagree with recommendations, which easily allowed for user corrections.	+
-	It worked for games that had enough interested members to enter recommendations, but didn't work for games with not enough interested members. This was the reason why a computerized algorithm was suggested as a replacement - to provide recommendations for all games (even though it still has a threshold requirement).	+