ETHIO PLANET – Oct 8 – When the Netflix Prize was awarded last month, it ended three years of intense competition aimed at finding a better algorithm for predicting users’ movie preferences. The winning team, BellKor’s Pragmatic Chaos, was the first to forecast Netflix customers’ movie ratings with 10% better accuracy than the company’s in-house system. Gavin Potter, who gained recognition for his breaking the top 10 of the Netflix prize in 2008, is working on applying his own research for the prize to the dating site YesNoMayB, which employs two-way recommendation algorithms to find users who may want to meet one another. In particular, he hopes to use insights from the Netflix Prize to make predictions based on users’ implicit preferences, such as what pages they load.
The full article was originally published at Ethio Planet, but is no longer available.
Online Dating Sites can be classified as:
Online Dating 1.0: First Generation “Browsing/Searching Options, Powerful Searching Engine”
Online Dating 1.5: Hybrid; “Unidirectional Recommendation Engine”, sites like HotorNot.
Online Dating 2.0: Second Generation “Matching based on Self-Reported Data / Bidirectional Recommendation Engine” e.g. PerfectMatch, uses an ipsative instrument based on MBTI test.
Online Dating 3.0: Third Generation “Compatibility Matching Algorithms” e.g. eHarmony, uses a normative version of the Big5 to assess personality and Dyadic Adjustment Scale to calculate compatibility. The U.S. questionnaire is different from the UK site, Australian site, Canadian site, etc.
Mutual filtering method (from Bidirectional Recommendation Engines) is in the range of 3 or 4 persons highly compatible (who select to each other) per 1,000 persons screened!!! so in a 10,000,000 persons database, one person will see 30,000 to 40,000 persons as highly compatible; 30,000 persons is the population of an average small city!!!
You can also see the paper: “Recommender System for Online Dating Service” Lukas Brozovsky & Vaclav Petricek (2007) showing:
“User-User and Item-Item collaborative filtering recommenders significantly outperform global algorithms that are currently used by dating sites [offering only Browsing / Searching Options, Powerful Searching Engine but not Compatibility Matching Algorithms].
A blind experiment with real users [at a proprietary site named ColFi – exclusively designed for the experiment – where 111 users rated 150 photo-profiles, then two recommendation lists of top 10 profiles were generated] also confirmed that users prefer collaborative filtering based recommendations to global popularity recommendations [of 2 Czech online dating sites: ChceteMe (no longer exists now in 2009) and LibimSeTi].”
“Recommendations can be further improved by hybrid algorithms. These algorithms are combining the collaborative filtering approach with content information. Another problem specific to dating is that A_likes_B does not imply B_likes_A. Therefore each user should be probably presented with recommendations of such users, who are also interested in him/her. There is a need for reciprocal matching algorithms.
User interface may introduce bias in the sense that users instead of providing their personal preference try to guess the global preference. This reduces the usefulness of ratings provided.”
[Vaclav Petricek is Sr. Matching Researcher at eHarmonyLabs since August 2007]
http://www.occamslab.com/petricek/papers/dating/brozovsky07recommender.pdf
Bidirectional Recommendation Engines do not take into account the new discovery uncovered by Eastwick and Finkel 2008; also Kurzban and Weeden, 2007; Todd, Penke, Fasolo, and Lenton, 2007 who found that people often report partner preferences that are not compatible with their choices in real life.
AND
Latest Research in Theories of Romantic Relationships Development outlines: compatibility is all about a high level on personality similarity between prospective mates for long term mating with commitment.
Kindest Regards.
Fernando Ardenghi.
Buenos Aires.
Argentina.
ardenghifer@gmail.com
Online Dating Sites can be classified as:
Online Dating 1.0: First Generation “Browsing/Searching Options, Powerful Searching Engine”
Online Dating 1.5: Hybrid; “Unidirectional Recommendation Engine”, sites like HotorNot.
Online Dating 2.0: Second Generation “Matching based on Self-Reported Data / Bidirectional Recommendation Engine” e.g. PerfectMatch, uses an ipsative instrument based on MBTI test.
Online Dating 3.0: Third Generation “Compatibility Matching Algorithms” e.g. eHarmony, uses a normative version of the Big5 to assess personality and Dyadic Adjustment Scale to calculate compatibility. The U.S. questionnaire is different from the UK site, Australian site, Canadian site, etc.
Mutual filtering method (from Bidirectional Recommendation Engines) is in the range of 3 or 4 persons highly compatible (who select to each other) per 1,000 persons screened!!! so in a 10,000,000 persons database, one person will see 30,000 to 40,000 persons as highly compatible; 30,000 persons is the population of an average small city!!!
You can also see the paper: “Recommender System for Online Dating Service” Lukas Brozovsky & Vaclav Petricek (2007) showing:
“User-User and Item-Item collaborative filtering recommenders significantly outperform global algorithms that are currently used by dating sites [offering only Browsing / Searching Options, Powerful Searching Engine but not Compatibility Matching Algorithms].
A blind experiment with real users [at a proprietary site named ColFi – exclusively designed for the experiment – where 111 users rated 150 photo-profiles, then two recommendation lists of top 10 profiles were generated] also confirmed that users prefer collaborative filtering based recommendations to global popularity recommendations [of 2 Czech online dating sites: ChceteMe (no longer exists now in 2009) and LibimSeTi].”
“Recommendations can be further improved by hybrid algorithms. These algorithms are combining the collaborative filtering approach with content information. Another problem specific to dating is that A_likes_B does not imply B_likes_A. Therefore each user should be probably presented with recommendations of such users, who are also interested in him/her. There is a need for reciprocal matching algorithms.
User interface may introduce bias in the sense that users instead of providing their personal preference try to guess the global preference. This reduces the usefulness of ratings provided.”
[Vaclav Petricek is Sr. Matching Researcher at eHarmonyLabs since August 2007]
http://www.occamslab.com/petricek/papers/dating/brozovsky07recommender.pdf
Bidirectional Recommendation Engines do not take into account the new discovery uncovered by Eastwick and Finkel 2008; also Kurzban and Weeden, 2007; Todd, Penke, Fasolo, and Lenton, 2007 who found that people often report partner preferences that are not compatible with their choices in real life.
AND
Latest Research in Theories of Romantic Relationships Development outlines: compatibility is all about a high level on personality similarity between prospective mates for long term mating with commitment.
Kindest Regards.
Fernando Ardenghi.
Buenos Aires.
Argentina.
ardenghifer@gmail.com
Mark/Fernando, the comment above can’t be read because of something going on with lines of text that exceed the available width.
More saliently, great that the Netflix approach is being applied to the idea of compatibility testing. One of the most exciting things about the Netflix project was that it involved people who didn’t necessarily have pre-existing knowledge of why certain people would prefer certain movies, but who brought a completely fresh approach to the challenge. Perhaps it will help compatibility research think more out of the box.
The other sweet thing about the Netflix exercise was that everyone’s work got shared, so people were able to build on each others’ findings. The accuracy got better and better. It took a long time, but the problem was finally cracked.
Can’t wait to see what happens when some fresh thinking gets applied.
Jon
Mark/Fernando, the comment above can’t be read because of something going on with lines of text that exceed the available width.
More saliently, great that the Netflix approach is being applied to the idea of compatibility testing. One of the most exciting things about the Netflix project was that it involved people who didn’t necessarily have pre-existing knowledge of why certain people would prefer certain movies, but who brought a completely fresh approach to the challenge. Perhaps it will help compatibility research think more out of the box.
The other sweet thing about the Netflix exercise was that everyone’s work got shared, so people were able to build on each others’ findings. The accuracy got better and better. It took a long time, but the problem was finally cracked.
Can’t wait to see what happens when some fresh thinking gets applied.
Jon
Gavin Potter has been invaluable and very accurate in calculating similarities and 2-way recommendations on yesnomayB.com.
We are now teaming up via introAnalytics.com to share our findings with other dating sites and social media. The larger the site, the more data, the more beneficial to its members.
Allowing online daters to “discover” rather than “search” is probably online dating’s biggest user-centric innovation since 1999.
Gavin Potter has been invaluable and very accurate in calculating similarities and 2-way recommendations on yesnomayB.com.
We are now teaming up via introAnalytics.com to share our findings with other dating sites and social media. The larger the site, the more data, the more beneficial to its members.
Allowing online daters to “discover” rather than “search” is probably online dating’s biggest user-centric innovation since 1999.
Hi Mr. Tsinonis:
Do you have any Scientific Paper or White Paper?
Because in a previous interview, last April 11 2009, Mr. Potter had said:
“It is true, however, and slightly to our surprise that after the location of the potential date, attractiveness appears to be the most important characteristic regardless of the information provided. We’ve now worked the algorithms on dating sites that contain substantial profile information as well as websites that contain very little and this finding doesn’t seem to change.”
“attractiveness appears to be the most important characteristic regardless of the information provided” …. clearly means that online dating site is only for fun, for flirting, for entertainment purposes, for instant gratification and not for serious daters looking for a long term relationship with commitment.
Alexa says yesnomayB has no traffic.
Kindest Regards.
Fernando Ardenghi.
Buenos Aires.
Argentina.
ardenghifer@gmail.com
Hi Mr. Tsinonis:
Do you have any Scientific Paper or White Paper?
Because in a previous interview, last April 11 2009, Mr. Potter had said:
“It is true, however, and slightly to our surprise that after the location of the potential date, attractiveness appears to be the most important characteristic regardless of the information provided. We’ve now worked the algorithms on dating sites that contain substantial profile information as well as websites that contain very little and this finding doesn’t seem to change.”
“attractiveness appears to be the most important characteristic regardless of the information provided” …. clearly means that online dating site is only for fun, for flirting, for entertainment purposes, for instant gratification and not for serious daters looking for a long term relationship with commitment.
Alexa says yesnomayB has no traffic.
Kindest Regards.
Fernando Ardenghi.
Buenos Aires.
Argentina.
ardenghifer@gmail.com