=Paper= {{Paper |id=Vol-1922/paper4 |storemode=property |title=Personality Traits and Music Genre Preferences: How Music Taste Varies Over Age Groups |pdfUrl=https://ceur-ws.org/Vol-1922/paper4.pdf |volume=Vol-1922 |authors=Bruce Ferwerda,Marko Tkalcic,Markus Schedl |dblpUrl=https://dblp.org/rec/conf/recsys/FerwerdaTS17 }} ==Personality Traits and Music Genre Preferences: How Music Taste Varies Over Age Groups== https://ceur-ws.org/Vol-1922/paper4.pdf

Personality Traits and Music Genre Preferences: How Music
Taste Varies Over Age Groups
Bruce Ferwerda∗ Marko Tkalcic Markus Schedl
School of Engineering Faculty of Computer Science Dep. of Computational Perception
Jönköping University Free University of Bozen-Bolzano Johannes Kepler University
P.O. Box 1026 Piazza Domenicani 3 Altenberger Strasse 69
SE-551 11, Jönköping, Sweden I-39100, Bozen-Bolzano, Italy 4040, Linz, Austria
bruce.ferwerda@ju.se marko.tkalcic@unibz.it markus.schedl@jku.at
ABSTRACT health [17, 26], education [2, 21], movies [3], music [8–10, 14, 27]).
Personality traits are increasingly being incorporated in systems These studies normally analyze their sample as a whole and do
to provide a personalized experience to the user. Current work not consider differences based on age groups. Arnett [1] showed
focusing on identifying the relationship between personality and that especially those in their adolescence and emerging adulthood
behavior, preferences, and needs often do not take into account phases experience a heightened chance of ”storm and stress” 1 in
differences between age groups. With music playing an important which they try to find their place in society. Hence, differences
role in our lives, differences between age groups may be especially may occur in behavior, preferences, and needs throughout different
prevalent. In this work we investigate whether differences exist phases in life.
in music listening behavior between age groups. We analyzed a To investigate the relationship between personality and music
dataset with the music listening histories and personality infor- genre preferences over different age groups, we used a subset of
mation of 1415 users. Our results show agreements with prior the myPersonality dataset. Next to users’ personality scores, this
work that identified personality-based music listening preferences. subset consist of the listening history of Last.fm (an online music
However, our results show that the agreements we found are in streaming service) 2 users. By analyzing the listening histories
some cases divided over different age groups, whereas in other of 1068 users in relation to their personality and age, we found
cases additional correlations were found within age groups. With important differences across age groups. Our insights may help
our results personality-based systems can provide better music to inform personalized music systems. For example, personality-
recommendations that is in line with the user’s age. based music recommender systems can improve their cold-start
recommendations (e.g., [5, 28]) by better knowing which music
CCS CONCEPTS genres to recommend to their users of different age groups.
•Information systems → Recommender systems; •Human-centered
computing → User models; User studies;
2 RELATED WORK
Currently, there are two different personality related research di-
KEYWORDS rections focusing on: 1) personality-based personalization (e.g.,
health [17, 26], education [2, 21], movies [3], music [8–10, 14, 27])
Music, Personality, Recommender Systems, User Modeling, Age
and 2) implicit personality acquisition from user-generated content
Differences
(e.g., Facebook [12, 16], Twitter [22], Instagram [11, 13], and fusing
information [25]). For example, in the area of personality-based
1 INTRODUCTION personalization Ferwerda et al. [15] looked at differences in how
Personality has shown to be a stable construct over time, and re- users browse for music (i.e., browsing music by genre, activity, or
flects the coherent patterning of one’s affect, cognition, and desires mood) in an online music streaming service. Chen, Wu, & He [3]
(goals) as it leads to behavior [24]. This stability and coherency of investigated diversity preferences in movie recommendations. In
personality has shown to be useful for systems to infer users’ pref- the area of implicit personality acquisition research mainly focuses
erences and to provide personalized experiences to users (e.g., [8]). on user-generated content of users’ social media accounts. Quer-
Hu & Pu [19] showed that personality-based personalized systems cia et al. [22] found that how users behave on Twitter consist of
have an advantage over personalized systems not incorporating cues to predict their personality. Similarly, Golbeck, Robles, &
personality information in terms of increased users’ loyalty towards Turner [16] were able to develop a personality predictor based on
the system and decreased cognitive effort. the characteristics of a user’s Facebook account.
The relationships between personality traits and users’ behavior Current personality-based research does not take into account
preferences and needs are increasingly being investigated (e.g., differences between age groups. However, Arnett [1] notes that es-
pecially those in their adolescence and emerging adulthood phases
∗ Also affiliated with the Department of Computational Perception, Johannes Kepler
may show deviant behavior. With music been shown to play an
University, Altenberger Strasse 69, 4040, Linz (Austria), bruce.ferwerda@jku.at.
important role in our lives by providing support for a whole range
Workshop on Temporal Reasoning in Recommender Systems (RecTemp) at the 11th
1 Storm-and-stress is a term first coined by Hall [18] to refer to a period in life in which
ACM Conference on Recommender Systems (RecSys). August 31, 2017, Como, Italy.
Copyright ©2017 for this paper by its authors. Copying permitted for private and people experience turmoil and difficulties.
academic purposes 2 http://www.last.fm/
of daily activities we engage in (e.g., sports, studying, sleeping) [23], 12-19 20-39 40-65
differences (e.g., listening behavior, preferences, and needs) across 0.35
age groups may especially be prevalent. 0.3
0.25
In this work we analyze a dataset of an online music streaming 0.2
service consisting of the total listening history of their users. With 0.15
0.1
this dataset we investigate whether differences in music listening 0.05
behavior exist. 0

3 METHOD
In order to investigate the relationship between personality and mu-
sic genre preferences across age groups in an online music stream-
ing service, we made use of the myPersonality dataset. 3 The dataset
Figure 1: Normalized genre play-counts by age group.
originates from a popular Facebook application (”myPersonality”)
that is able to record psychological and Facebook profiles of users
that used the application to take psychometric (e.g., personality, at-
titudes, skills) tests. The dataset contains over 6 million test results, 4 RESULTS
with over 4 million Facebook profiles. Users’ personality in the For the analysis we filtered out users with zero play-counts (users
myPersonality application was assessed using the Big Five Inven- who registered, but did not make use of Last.fm) and people listen-
tory to measure the constructs of the five factor model: openness, ing to less than five different artists. This left us with a total of 1068
conscientiousness, extraversion, agreeableness, and neuroticism. users (adolescence: n =472, young adulthood: n =563, middle adult-
We only used the subset of the myPersonality dataset that con- hood: n =33). The normalized genre play-counts by the different
tains the music listening history of Last.fm users (i.e., play-count age groups are shown in Figure 1.
of artists that a user listened to) and the day of birth in order to To investigate music listening differences between personality
calculate their age. The subset consists of users’ complete listen- traits, Spearman’s correlation was computed between personality
ing histories (i.e., from the moment they started to use Last.fm) traits and the genre play-count to assess the relationship of person-
until April 27 (2012). We complemented the dataset by adding the ality and genre preferences. Alpha levels were adjusted using the
listening events of each user until December 18 (2016) by using Bonferroni correction to limit the chance on a Type I error. The
the Last.fm API. 4 A total of 1066 Last.fm users with ∼40 million reported significant results adhere to alpha levels of p <.001 (see
listening events from 101 countries are represented in the subset. Table 1). The results of music genre preferences per personality
The 1066 Last.fm users were split into three different age groups trait per age groups are discussed in the following sections.
according to the primary life stages [4]: adolescence (age: 12-19),
young adulthood (age: 20-39), and middle adulthood (age: 40-65). 4.1 Openness
Having the day of birth of the 1066 Last.fm users as well as their
complete listening history (with listening date), we could traceback Adolescence (age: 12-19). For the adolescence group, several pos-
users’ age when listened to a certain song. Hence, users could fall itive correlations were found with music genres: New Age (r =.142),
into multiple age groups, which resulted in a sample size bigger Blues (r =.130), Country (r =.117), World (r =.114), Folk (r =.230),
than the original sample. The final dataset consists of 1479 Last.fm Jazz (r =.139), Vocal (r =.132), and Alternative (r =.131). Those
users divided over three age groups (adolescence: n =581, young scoring high on the openness scale show in general higher listening
adulthood: n =850, middle adulthood: n =48). behavior to these music genres.
Through the Last.fm API, we crawled additional information
about the artists by using the ”Artist.getTopTags” endpoint. This Young adulthood (age: 20-39). For the young adulthood group,
endpoint provided us with all the tags that users assigned to an we found the same kind of positive correlations with music gen-
artist, such as instruments (“guitar”), epochs (“80s”), places (“Chicago”), res: New Age (r =.105), Blues (r =.167), Country (r =.126), World
languages (“Swedish”), and personal opinions (“seen live” or “my (r =.217), Folk (r =.231), Jazz (r =.106), Vocal (r =.170), and Alter-
favorite”). Tags that encode genre or style information were filtered native (r =.116). An additional positive correlation was found with
for each artist. The filtered tags were then indexed by a dictionary Electronic (r =.106), and a negative correlation with Rock (r =-.104)
of 18 genre names retrieved from Allmusic. 5 For each user in an indicating that those scoring high on openness tend to listen less
age group, the artists that were listened to were aggregated by to Rock music in this age group. Next to Folk music, a stronger
the indexed genre with their play-count. The genre play-count for correlation was found for World music as well.
each user was then normalized to represent a range of rϵ[0,1], this
in order to be able to compare users with differences in the total Middle adulthood (age: 40-65). When those scoring high on
amount of listening events. openness reach middle adulthood, their variation of listening to mu-
sic genres shrinks significantly, but the strength of the correlations
increases. We found positive correlations with Blues (r =.358) as
3 http://mypersonality.org/ well as with Folk (r =.368) music. Indicating that although the vari-
4 http://www.last.fm/api ation gets less, the music genre preferences for middle adulthood
5 http://www.allmusic.com becomes more prominent.
Openness Conscientiousness Extraversion Agreeableness Neuroticism
12-19 20-39 40-65 12-19 20-39 40-65 12-19 20-39 40-65 12-19 20-39 40-65 12-19 20-39 40-65
R&B -.019 -.004 -.053 -.026 -.009 .150 .106 .065 .326 -.049 .047 .326 .027 -.001 -.175
Rap -.019 -.011 -.205 -.085 -.065 .059 .030 .108 .052 -.070 .062 .052 .003 -.072 -.158
Electronic .046 .106 -.138 -.043 -.031 .152 .015 .038 -.246 -.090 -.050 -.246 .036 -.023 .133
Rock -.075 -.104 .095 -.058 .016 -.124 -.085 -.102 -.182 .070 -.031 -.182 .014 .053 .182
New Age .142 .105 .133 .037 -.053 .006 -.022 -.184 -.209 .008 .011 -.209 -.062 -.064 -.143
Classical .080 .038 .266 .028 -.060 .261 -.136 -.146 -.136 -.070 -.010 -.136 -.015 -.005 -.080
Reggae -.015 .046 .185 -.102 -.059 -.059 .039 .025 .046 -.032 .051 .046 .028 -.042 -.138
Blues .130 .167 .358 -.048 -.046 .321 .060 .032 .252 -.006 .018 .252 -.054 -.005 -.552
Country .117 .126 .325 -.067 -.073 .154 .005 .005 .128 .062 .184 .128 .049 -.027 -.109
World .114 .217 .201 -.016 -.009 .217 -.102 -.054 .028 -.056 -.025 .028 .061 -.014 -.236
Folk .230 .231 .368 -.014 -.114 -.268 .066 -.040 .181 .101 .110 .181 -.064 .004 -.217
Easy
.084 .060 -.161 .020 .024 .256 .041 -.019 .212 -.073 .041 .212 .035 -.012 .006
Listening
Jazz .139 .106 -.124 -.047 -.025 .510 .005 -.010 .062 -.053 -.068 .062 -.039 .004 -.106
Vocal (a
.132 .170 .282 .059 -.007 .125 .038 -.013 .136 -.074 -.001 .136 -.014 .002 -.091
cappella)
Punk -.032 -.008 .089 -.130 -.103 .081 -.111 -.029 -.074 .005 .006 -.074 .101 .049 .220
Alternative .131 .116 .154 -.108 -.165 .507 -.010 -.052 -.027 .018 .029 -.027 .129 .137 .070
Pop .021 .000 -.157 .045 .005 .052 .064 .017 .287 -.017 .194 .287 .040 -.010 -.275
Heavy
-.033 -.044 -.117 -.005 -.012 .038 -.148 -.126 -.339 -.058 -.105 -.339 -.030 -.030 .372
Metal
Table 1: Spearman’s correlation between music genres and personality traits over age groups. Significant correlations after
Bonferroni correction are shown in boldface (p <.001).

4.2 Conscientiousness (r =-.146), and Heavy Metal (r =-.126). A positive correlation was
Adolescence (age: 12-19). Those scoring high on conscientious- found with Rap (r =.108) music.
ness in the adolescence group show mainly negative correlations
with the listened music genres: Reggae (r =-.102), Punk (r =-.130), Middle adulthood (age: 40-65). The middle adulthood group
and Alternative (r =-.108). The results indicates that for this age of the extraverts show a positive correlation with R&B (r =.326)
group, conscientious users listen less to these music genres. and a negative correlation with Heavy Metal (r =-.339).

Young adulthood (age: 20-39). Negative correlations were also 4.4 Agreeableness
found between music genres and conscientiousness for the young Adolescence (age: 12-19). The adolescence group show only a
adulthood group. Although the Punk (r =-.103) and Alternative positive correlation with agreeableness for Folk (r =.101) music.
(r =-.108) music genre is in line with the adolescence group, instead This indicates that agreeable users show in this age group show a
of Reggae, this group listens less to Folk (r =-.114) music. preference for Folk music.

Middle adulthood (age: 40-65). We found two correlations for Young adulthood (age: 20-39). A more varied music preference
the middle adulthood group: Jazz (r =.510) and Alternative (r =.507). is shown for the young adulthood group. Positive correlations were
Both correlation coefficients show high effects between conscien- found for Country (r =.184), Folk (r =.110), and Pop (r =.194) music.
tiousness and the music genres. A negative correlation was found for Heavy Metal (r =-.105) music.
Agreeable users in their young adulthood phase seem to prefer to
4.3 Extraversion listen to Country, Folk, and Pop, but less to Heavy Metal music.
Adolescence (age: 12-19). The adolescence group show nega-
tive correlations with Classical (r =-.136), World (r =-.102), Punk Middle adulthood (age: 40-65). The middle adulthood group
(r =-.111), and Metal (r =-.148). A positive correlation was found show a negative correlation with Heavy Metal (r =-.339) music,
with R&B (r =.106). The results indicate that extraverts in their which indicates that their preference to listen to Heavy Metal goes
adolescence phase listen less to Classical, World, Punk, and Heavy down when reaching the age of middle adulthood.
Metal music. However, the in general listen to more R&B.

Young adulthood (age: 20-39). For those scoring high on ex-
traversion and fall in the young adulthood group show negative
correlations with Rock (r =-.102), New Age (r =-.184), Classical
4.5 Neuroticism music is positively correlated with neuroticism for adolescence, but
Adolescence (age: 12-19). Neurotics in their adolescence phase only Alternative music is positively correlated with neuroticism
show positive correlations with Punk (r =.101) as well as with in the young adulthood group. Moreover, we show only a positive
Alternative (r =.129) music, indicating an increase preference for correlation with Heavy Metal in the middle adolescence group.
these music genres.
6 CONCLUSION & IMPLICATIONS
Young adulthood (age: 20-39). Only a positive correlation with In this work we investigated whether there are differences across
Alternative (r =.137) was found in the young adulthood group. age groups in the relationship between personality traits and music
genre preferences. When not considering differences across age
Middle adulthood (age: 40-65). For the middle adulthood group, groups, we show that we found agreements with prior works [20,
music preferences seem to switch. A positive correlation was found 23] on the relationship between personality and music genre pref-
with Heavy Metal (r =.372) and a negative correlation was found erences. Whereas prior works analyzed their sample as a whole,
with Blues (r =-.552). we show with our results that differences exist in music genre pref-
erences depending on age groups. With our results we validate
5 DISCUSSION the results of prior works, but show that there are cases where the
previously found correlations with music preferences are divided
Our results show that there are differences in music listening be- over different age groups, whereas in other cases other (previously
havior between personality traits, and that these difference can unrevealed) correlations show up within age groups.
be further broken down by age groups. Overall, our results show Our work contributes to the personality-based work for per-
that users in their adolescence and young adulthood phases show sonalized systems. The differences between age groups that we
most variation in their music listening behavior. Not only does identified in this work may have important implications for the cre-
the variation become much less when reaching middle adulthood, ation of personalized systems. The focus of the recommendations
the correlation strength increase significantly. This indicates that may differ depending on the age groups a user falls in. For example,
music preferences of the middle adulthood group becomes more the recommendations for adolescent extraverts could be focused
established, which is in line with the storm-and-stress argument [1]. on R&B music, whereas recommendations for extraverts in their
The openness trait shows most variation in listening to different young adulthood could be more focused on Rap music.
music genres amongst the personality traits. This is in line with one For our future work, we will extend our findings by actually
of the few works that investigated the relationship between person- trying to provide music recommendations to users and perform
ality traits and music listening behavior [23]. However, what their a user-centric evaluation on the recommendations. For example,
findings do not show is that there are differences when considering including diversity in recommendations have shown to be an impor-
age groups. For example, the addition of a preference for Electronic tant feature on satisfaction [29]. In addition, Ferwerda et al. [6, 7]
music in the young adulthood group. identified the prerequisites for diversification and found differences
Also the conscientiousness trait shows agreement with prior in diversity needs among personality traits. Our findings could help
work [20]. However, additional unique correlations were able to to inform the diversification in recommendations by incorporating
be identified when taking different age groups into account. Our different needs across age groups. For example, those scoring high
results show that the adolescence group shows an additional nega- on openness in their adolescence or young adulthood phases may
tive correlation with Reggae, the young adulthood group shows an be given more diverse genres (correlations were found with eight
additional correlation with Folk music, and the middle adulthood and ten different genres respectively), whereas the recommenda-
group shows an additional positive correlation with Jazz music. tions for those in their middle adulthood can be narrowed down to
Our results on extraversion show agreements with prior works [20, Blues and Folk.
23] as well. However, what the results of prior works do not show In this work, we also did not take into account possible cultural
is that there is a division based on age. For example, our results differences. Although having the music listening histories of users
show that the positive correlation of R&B and Rap, differ across age from different countries, we disregarded country information in
groups. The adolescence and middle adulthood group show positive order to keep a big enough sample. In future work we will address
correlations only with R&B, whereas only a positive correlation possible cultural differences.
with Rap was found with the young adulthood group.
For the agreeableness trait, we found agreements with prior
work [23] especially for the young adulthood group show: positive 7 ACKNOWLEDGMENTS
correlations with Country, Folk, and Pop music. These full agree- Supported by the Austrian Science Fund (FWF): P25655. We would
ments seem to only hold for the young adulthood group. We found also like to thank Michal Kosinski and David Stillwell of the myPer-
less agreements with the adolescence and the middle adulthood sonality project for sharing the data with us.
group: only Folk music showed to be positively correlated.
The agreements we found with prior work [20] on neuroti- REFERENCES
cism are divided across age groups. Whereas prior works showed [1] Jeffrey Jensen Arnett. 1999. Adolescent storm and stress, reconsidered. American
grouped correlations with Punk, Alternative, and Heavy Metal psychologist 54, 5 (1999), 317.
[2] Guanliang Chen, Dan Davis, Claudia Hauff, and Geert-Jan Houben. 2016. On
music on neuroticism, our results show that these correlations do the impact of personality in massive open online learning. In Proceedings of the
not hold for all age groups. We show that Punk and Alternative 2016 conference on user modeling adaptation and personalization. ACM, 121–130.
[3] Li Chen, Wen Wu, and Liang He. 2013. How personality influences users’ needs systems. ACM, 253–262.
for recommendation diversity?. In CHI’13 Extended Abstracts on Human Factors [17] Sajanee Halko and Julie A Kientz. 2010. Personality and persuasive technology:
in Computing Systems. ACM, 829–834. an exploratory study on health-promoting mobile applications. In International
[4] Erik H Erikson. 1993. Childhood and society. WW Norton & Company. Conference on Persuasive Technology. Springer, 150–161.
[5] Ignacio Fernández-Tobı́as, Matthias Braunhofer, Mehdi Elahi, Francesco Ricci, [18] Granville Stanley Hall. 1916. Adolescence: Its psychology and its relations to
and Iván Cantador. 2016. Alleviating the new user problem in collaborative physiology, anthropology, sociology, sex, crime, religion and education. Vol. 2. D.
filtering by exploiting personality information. User Modeling and User-Adapted Appleton.
Interaction 26, 2-3 (2016), 221–255. [19] Rong Hu and Pearl Pu. 2009. Acceptance issues of personality-based recom-
[6] Bruce Ferwerda, Mark Graus, Andreu Vall, Marko Tkalcic, and Markus Schedl. mender systems. In Proceedings of the third ACM conference on Recommender
2016. The influence of users’ personality traits on satisfaction and attractiveness systems. ACM, 221–224.
of diversified recommendation lists. In 4 th Workshop on Emotions and Personality [20] Alexandra Langmeyer, Angelika Guglhör-Rudan, and Christian Tarnai. 2012.
in Personalized Systems (EMPIRE) 2016. 43. What do music preferences reveal about personality? Journal of Individual
[7] Bruce Ferwerda, Mark P Graus, Andreu Vall, Marko Tkalcic, and Markus Schedl. Differences (2012).
2017. How item discovery enabled by diversity leads to increased recommenda- [21] Michael J Lee and Bruce Ferwerda. 2017. Personalizing online educational tools.
tion list attractiveness. In Proceedings of the Symposium on Applied Computing. In Proceedings of the 2017 ACM Workshop on Theory-Informed User Modeling for
ACM, 1693–1696. Tailoring and Personalizing Interfaces. ACM, 27–30.
[8] Bruce Ferwerda and Markus Schedl. 2014. Enhancing Music Recommender [22] Daniele Quercia, Michal Kosinski, David Stillwell, and Jon Crowcroft. 2011. Our
Systems with Personality Information and Emotional States: A Proposal.. In Twitter profiles, our selves: Predicting personality with Twitter. In Proceedings
UMAP Workshops. of the International Conference on Social Computing (SocialCom). IEEE, 180–185.
[9] Bruce Ferwerda and Markus Schedl. 2016. Personality-Based User Modeling for [23] Peter J Rentfrow and Samuel D Gosling. 2003. The do re mi’s of everyday life: the
Music Recommender Systems. In Joint European Conference on Machine Learning structure and personality correlates of music preferences. Journal of personality
and Knowledge Discovery in Databases. Springer, 254–257. and social psychology 84, 6 (2003), 1236.
[10] Bruce Ferwerda, Markus Schedl, and Marko Tkalcic. 2015. Personality & Emo- [24] William Revelle. 2009. Personality structure and measurement: The contributions
tional States: Understanding Users’ Music Listening Needs.. In UMAP Workshops. of Raymond Cattell. British Journal of Psychology 100, S1 (2009), 253–257.
[11] Bruce Ferwerda, Markus Schedl, and Marko Tkalcic. 2015. Predicting personality [25] Marcin Skowron, Marko Tkalčič, Bruce Ferwerda, and Markus Schedl. 2016.
traits with instagram pictures. In Proceedings of the 3rd Workshop on Emotions Fusing social media cues: personality prediction from twitter and instagram. In
and Personality in Personalized Systems 2015. ACM, 7–10. Proceedings of the 25th international conference companion on world wide web.
[12] Bruce Ferwerda, Markus Schedl, and Marko Tkalcic. 2016. Personality traits and International World Wide Web Conferences Steering Committee, 107–108.
the relationship with (non-) disclosure behavior on facebook. In Proceedings of [26] Kirsten A Smith, Matt Dennis, and Judith Masthoff. 2016. Personalizing reminders
the 25th International Conference Companion on World Wide Web. International to personality for melanoma self-checking. In Proceedings of the 2016 Conference
World Wide Web Conferences Steering Committee, 565–568. on User Modeling Adaptation and Personalization. ACM, 85–93.
[13] Bruce Ferwerda, Markus Schedl, and Marko Tkalcic. 2016. Using instagram [27] Marko Tkalčič, Bruce Ferwerda, David Hauger, and Markus Schedl. 2015. Per-
picture features to predict usersfi personality. In International Conference on sonality correlates for digital concert program notes. In International Conference
Multimedia Modeling. Springer, 850–861. on User Modeling, Adaptation, and Personalization. Springer, 364–369.
[14] Bruce Ferwerda, Marko Tkalcic, and Markus Schedl. 2017. Personality Traits and [28] Marko Tkalcic, Matevz Kunaver, Andrej Košir, and Jurij Tasic. 2011. Address-
Music Genres: What Do People Prefer to Listen To?. In Proceedings of the 25th ing the new user problem with a personality based user similarity measure. In
Conference on User Modeling, Adaptation and Personalization. ACM, 285–288. Proceedings of the 1st International Workshop on Decision Making and Recommen-
[15] Bruce Ferwerda, Emily Yang, Markus Schedl, and Marko Tkalcic. 2015. Personal- dation Acceptance Issues in Recommender Systems. Citeseer, 106.
ity traits predict music taxonomy preferences. In Proceedings of the 33rd Annual [29] Martijn C Willemsen, Bart P Knijnenburg, Mark P Graus, LC Velter-Bremmers,
ACM Conference Extended Abstracts on Human Factors in Computing Systems. and Kai Fu. 2011. Using latent features diversification to reduce choice difficulty
ACM, 2241–2246. in recommendation lists. RecSys 11 (2011), 14–20.
[16] Jennifer Golbeck, Cristina Robles, and Karen Turner. 2011. Predicting personality
with social media. In CHI’11 extended abstracts on human factors in computing