chi 2014 panel: opportunities and risks of discovering personality traits from social media
DESCRIPTION
We will conduct a panel at CHI 2014 discussing the opportunities and implications of the coming wave of new analytics that allow individuals' intrinsic traits, such as personality and motivations, to be mined from their behaviors on social platforms. For more information, see the Facebook page for the panel here: https://www.facebook.com/events/633305060096913/TRANSCRIPT
Interacting with Digital Individuals:Opportunities and Risks of Discovering Personality Traits from Social Media
ModeratorsMichelle Zhou, Jeffrey Nichols
PanelistsVictoria Bellotti, Tom Dignan, Jennifer Golbeck, Jeff Hancock
Michelle Zhou Jeffrey NicholsIBM Research – Almaden, IBM Watson Group
Outline
• Introduction (30 min)• Digital individuals from Social Media (live demo)
• Intro presentations from each Panelist
• Q&A with audience (45 min)
• Summary (5 min)
Analytics of Aggregates
Monitoring and Reporting
Analytics of Individuals
Sentiment
Listening Engagement Workflow
Measurement
Publishing
Net Promoter
Network Topology
Mas
s
Indivi
dual
Intrinsic Traits
What are people saying?
How do people feel about my brand?
Who is this individual? What motivates her? What is her taste and style?
Next generation
Earlystages
State of the art
Social Genome
• Demographics• Birthday• Age• Home Location• Political Affiliation• Religion• Etc.
• Intrinsic Traits• Personality (Big 5)• Basic Human Values
(Motivations)• Fundamental Needs
(Buying behaviors)• Emotional State• Etc.
Information That Can Be Extracted
From Social Media
DEMO
IBM System U
Methodology: Personality Analytics
“I love food, .., with … together we … in… very…happy.”
Word category: Inclusive Agreeableness
[Tausczik and Pennebaker ‘10, Yarkoni ‘10]
Automatically compute one’s personality traits
Make hyper-personalized recommendations based on derived traits[Ford ‘05, O’Brien ‘96, Neuman ‘99, Gosling ‘03, Wholan ‘06]
Do it for hundreds of millions of individuals
Opportunities Individualization at Scale
“Welcome to our store, would you like to take a personality test?”
Privacy invasion
Veracity of social media
Analytics imperfections
RisksIndividualization misfire
“Your tweets tell us that you appear to have multiple
personalities”
“We do not hire vulnerable people like you”
ProfessorCommunication, Information ScienceCornell University
Areas of Interest• Computer-mediated
communication• Language and technology• Deception and its detection • Figurative language
Jeff Hancock
ProfessorCollege of Information StudiesUniversity of Maryland
Areas of Interest• Trust modeling• Personality and political
preference from social media• Usable Security
Jennifer Golbeck
Vice President, Head of Research Reputation.com
Areas of Interest• Reputation scoring• Big data analytics and
platforms for real-world systems
Tom Dignan
Reputation.com is: The pioneering leader in the online reputation management &
digital privacy space We monitor the online presence of individuals and businesses
What shows up in your search results What people say about you on Facebook, Twitter, and other
social media sites What PPI do you have exposed What photos or videos are tagged with your name What public records are exposed
We assess and manage your digital footprint Is the content about you negative or positive (sentiment
analysis) How much personal information do you have exposed Where is your online presence lacking
What Does Reputation.com Do?
The Driving Factors for Reputation.com’s Business:
As more and more of people’s lives are conducted online, there is a growing desire to enjoy the benefits of interacting on the internet whilst maintaining control of one’s online profile, presence and personal data.
Philosophy: people and businesses have a right to control, protect their online reputations and privacy.
Online reputation will only grow in importance: ample research demonstrates that consumers are actively searching online and trusting what they find – and businesses are materially impacted by social media/review feedback.
What is Driving Reputation.com’s Business?
Research FellowPARC
Areas of Interest• Ethnography• Task and activity management• Context-aware computing• Sharing economy and
collaborative consumption
Victoria Bellotti
Context-Aware Computing: Activity Spotting
Time
No. of crowdsourced
activity features
spotted per interval
Inferential threshold Writing a paper
Not writing a paper
Ben Sutherland - Flickr
Rethinking the TimeBanking Metaphor: Do humans need to be paid to be nice?
Policies and Practice: Doing the Right ThingWednesday 2:00-3:30pm Room: 801BTalk is scheduled to start at 3:10 pm
Q&A
• How to interpret or measure the accuracy of personality traits derived from social media?
• What factors (e.g., data sources and analytics methods) may affect accuracy?
• What could the derived personality traits be used for individuals and businesses?
• Who would benefit the most? • What are the risks of using such technologies, especially from an
individual perspective? • Who is at risk?• How can we protect ourselves?• …
Summary
• To be filled in during the panel
Validation
How good are our results compared to standard psychometric studies?
How well can our results be used to predict or influence one’s behavior?
Yarkoni ’10, Adali ‘12, Chen ‘14, Gou ’14 …
Mahmud ‘13, Lee ‘14
Results
• RV-Coefficient correlation analysis of each type of trait
• Over 80% of population, their correlation is statistically significant (80.8%, 98.21%, and 86.6% for Big 5 personality, basic values and needs)
[Gou et al. CHI 2014]
References• Chen, J., Hsieh, G., Mahmud, J., and Nichols, J. Understanding individuals personal values from social media
word use. In ACM Proc. CSCW ’2014. • Ford, J. K. Brands Laid Bare. John Wiley & Sons, 2005. • Gou, L., Zhou, M.X., and Yang, H. KnowMe and ShareMe: Understanding automatically discovered personality
traits from social media and user sharing preferences. In ACM Proc. CHI 2014.• Lee, K., Mahmud, J., Chen, J., Zhou, M.X., and Nichols, J. Who will retweet this? Automatically identifying and
engaging strangers on Twitter to spread information. In ACM Proc. IUI ‘2014.• Luo, L., Wang, F., Zhou, M.X., Pan, X., and Chen, H. Who’s got answers? Growing the pool of answerers in a
smart enterprise Social Q&A system. In ACM Proc. IUI ‘2014. • Mahmud, J., Zhou, M.X., Megiddo, N., Nichols, J., and Drews, C. Recommending Targeted Strangers from Whom
to Solicit Information in Twitter. In ACM Proc. IUI ‘2013. • Schwartz, S. H. Basic human values: Theory, measurement, and applications. Revue francaise de sociologie,
2006. • Tausczik, Y. R., and Pennebaker, J. W. The psychological meaning of words: LIWC and computerized text analysis
methods. Journal of Language and Social Psychology 29, 1 (2010), 24–54.• Yang, H., and Li, Y. Identifying user needs from social media. IBM Tech. Report (2013).• Yarkoni, T. Personality in 100,000 words: A large-scale analysis of personality and word use among bloggers. J.
research in personality 44, 3 (2010), 363–373.