Tuesday, May 17, 2011

NEW and FRESH paper, punch to Behavioural Recommenders.

This NEW and FRESH paper is another big punch to Behavioural Recommender Systems / Recommendation Engines for Online Dating sites, like the ones used at Match or PlentyOfFish and the one Meetic plans to launch soon!!!
"Product innovation, the growth engine behind Meetic Group's development. ...These Apps will allow users to import and integrate into their profile their tastes in terms of music, films and literature, their adherence to various communities and other information related to their social network usage. This wealth of information will thus allow Meetic to further improve the way it puts users in contact with each other through a new Social Matching algorithm."

"How Shared Preferences in Music Create Bonds Between People: Values as the Missing Link" 2011
Diana Boer, Ronald Fischer, Micha Strack, Michael H. Bond, Eva Lo, and Jason Lam
How can shared music preferences create social bonds between people?
A process model is developed in which music preferences as value-expressive attitudes create social bonds via conveyed value similarity. The musical bonding model links two research streams: (a) music preferences as indicators of similarity in value orientations and (b) similarity in value orientations leading to social attraction. Two laboratory experiments and one dyadic field study demonstrated that music can create interpersonal bonds between young people because music preferences can be cues for similar or dissimilar value orientations, with similarity in values then contributing to social attraction. One study tested and ruled out an alternative explanation (via personality similarity), illuminating the differential impact of perceived value similarity versus personality similarity on social attraction. Value similarity is the missing link in explaining the musical bonding phenomenon [
and not personality similarity], which seems to hold for Western and non-Western samples and in experimental and natural settings.

Empirical evidence has shown that music preferences are associated with both personality traits and value orientations. Individuals who reject conservative values and who endorse openness to change values like listening to rock and punk; individuals who are guided by self-enhancing and openness values tend to like popular music, such as international pop and hip-hop; and individuals with self-transcendent value priorities like listening to jazz and classic. These associations seem to hold across Western and non-Western cultures and across value measurements.

Study 1
Participants were 338 German music fans of rock, metal, hip-hop, and electronic music (electro). Their average age was 17.80 (SD = 4.49), and 189 participants were female.
Study 1 supports the proposed musical bonding model: Shared music preferences can generate social attraction, and this effect is likely to be induced by assumed value similarity.
However, values and personality are related concepts and music taste is related to personality
traits. Hence, assumed similarity in personality traits, not accounted for in Study 1, might
be an alternative explanation for the musical bonding effect.
In Study 2, similarity in values and personality was directly measured instead of getting separate ratings of self-ratings and attributed ratings.
Participants were 67 German fans of metal and hip-hop music (age M = 21.15 years, SD =
6.35; n = 50 male participants, n = 36 metal fans).
Similarity in value orientations and personality traits between participants and targets was assessed using direct similarity assessments.
Value similarity and personality similarity showed good internal consistency (value similarity: Cronbach's alpha = .94; personality similarity: Cronbach's alpha = .93).
Study 3 examined musical bonding in a natural setting.
Participants were 94 undergraduate students at a university in Hong Kong (age M = 20.44, SD = 1.19; n = 34 female) who had been randomly allocated to dormitories by the university administration 1 to 2 months earlier.
Two individuals always share one room. The sample composed 47 same-sex roommate dyads (30 male, 17 female dyads). Participants answered a self-report survey including questions about their value orientations, music preferences, social attraction and perceived similarity of their roommate, and demographic details including age, gender, subject of study, and nationality.
Music preference similarity was calculated based on raw score ratings of music styles in the three factors: Western styles, Chinese classical styles, and pop styles (Pearson correlation).

We showed that value expression through music is likely to lead to social bonding but not similarity in basic personality traits.

This paper is not in the TOP 33 scientific papers for the Online Dating Industry. (a list I had sent you in a previous email)
Please add as number 34.

[Please remember:
Behavioural Matching recommends people based on the type of person you have sent emails to, replied to, clicked on in search results and gone on dates with, but .... persons/people often report/select partner preferences that are not compatible with their choices in real life -uncovered by Eastwick & Finkel (2008); Kurzban & Weeden (2007); Todd, Penke, Fasolo, & Lenton (2007)-.

Latest Research in Theories of Romantic Relationships Development shows: compatibility is all about a high level on personality* similarity* between prospective mates for long term mating with commitment
*personality measured with a normative test.
*similarity: there are different ways to calculate similarity, it depends on how mathematically is defined.
Also several studies showing contraceptive pills users make different mate choices, on average, compared to non-users.]


Also read
birds of a feather sing together (the 1998 paper)
birds of a feather sing together (the post at SignalPatterns Labs 2008)

The notion of music compatibility was pioneered by Dr. Jason Rentfrow and Dr. Sam Gosling, but never succeed.
"Several empirical studies in music psychology have identified connections between the styles of music people like listening to and a range of personality traits.
...
Another study concerned with the links between music preferences and relationship satisfaction focused on music-preference similarity among university roommates living in the US. The roommates who participated in the study completed several surveys, including a personality questionnaire, a music preference measure, and a few questions about how much they enjoyed living with their roommate and whether they would choose to live with them the following year. The results showed that roommates with similar personalities were no more likely to enjoy their living arrangement than were roommates with different personalities. However, roommates with similar music preferences enjoyed more pleasant and satisfying relationships and reported a stronger desire to continue living together than did roommates with different music preferences. Closer inspection of the results indicated that roommates with similar preferences for rap and dance music, and similar preferences for classical and jazz music were the most likely to enjoy their living situations. Furthermore, students tended to segregate themselves into dorms with other students that tended to match their music preferences than any of the other characteristics measured in the study.
...
Given that there are connections between the styles of music people listen to and their personalities, it is reasonable to suppose that people with similar music preferences may be more likely to get along and enjoy happy relationships than people with radically different preferences for music."

ALL RECOMMENDER SYSTEMS / RECOMMENDATION ENGINES ARE RUBBISH FOR SERIOUS DATERS.

No comments:

Post a Comment