Table Of ContentDesigning Personalization in
Technology-Based Services
by
Min Kyung Lee
September, 2013
CMU-HCII-13-102
Human-Computer Interaction Institute
School of Computer Science
Carnegie Mellon University
Pittsburgh, Pennsylvania 15213
Thesis Committee:
Jodi Forlizzi (Co-chair), Carnegie Mellon University
Sara Kiesler (Co-chair), Carnegie Mellon University
John Zimmerman, Carnegie Mellon University
Leila Takayama, Google
Submitted in partial fulfillment of the requirements
for the degree of Doctor of Philosophy
Copyright © 2013 Min Kyung Lee. All rights reserved.
This work was supported by the National Science Foundation (IIS-0121426, CRI-0709077),
Microsoft Research, and the author’s KwanJeong Lee Jong Hwan Scholarship and Siebel
Scholarship. Any opinions, findings, conclusions, or recommendations expressed in this material
are those of authors and do no necessarily reflect those of the funding agencies.
Keywords
Personalization, Technology-Based Services, Service Design, Human-Robot
Interaction, Human-Computer Interaction, Computer-Supported Cooperative
Work, Long-Term Interaction, Cooperation, Rapport, Engagement, Script,
Schema, Service Breakdowns, Recovery Strategies, Breakdown Mitigation
Strategy, Personal Service Providers, Adaptive Service Design, Co-Creation,
Human-Centered Design, Interaction Design
2
Acknowledgements
It takes ten years to have an idea really one’s own — about which one can talk.
The Wind at Djemila by Albert Camus
My passage to crafting my own idea and voice has not been a solitary individual
endeavor, but rather one accompanied by many mentors, friends, and family
members. My Ph.D. advisors, Sara Kiesler and Jodi Forlizzi, provided essential
guidance, support, and encouragement throughout my intellectual journey. Sara
Kiesler has exemplified how vigorous curiosity and open-mindedness can push
boundaries and transform diverse perspectives into beautiful, novel ideas. Jodi
Forlizzi has been an example of strength and determination, bringing design
principles into human-computer interaction and transforming design as an
academic field. Thank you for being my role models.
John Zimmerman, one of my thesis committee members, has prompted a critical
reflection on my work, honing my contribution and making it much more
interesting. He once offered a research opportunity to a budding student who had
just arrived at CMU. It was my first experience with research and played a critical
role in my decision to pursue the Ph.D. program. Thank you.
Leila Takayama, another thesis committee member, has encouraged me to strive
to make an impact on both theory and practice. She created a place for me at
Willow Garage; since then, she has been incredibly helpful and has shown me
what it means to be a powerful yet supportive leader. Thank you.
Several scholars, many of them not cited in my dissertation, deeply influenced my
intellectual growth and brought joy to my research career. George Loewenstein,
Richard Buchanan, Donald Schon, Wanda Orlikowski, Robert Sutton, Jennifer
Aaker, Daniel Kahneman, Amos Tversky, and Herbert Simon: Thank you.
3
All of my projects were possible thanks to great collaborators such as Siddhartha
Srinavasa, Paul Rybski, Reid Simmons, Pamela Hinds, Maxim Makatchev, Kyle
Strabalasa, and Anca Dragan. Being in the HCI Institute, I also had some
wonderful opportunities for interacting with and receive advice from many
bright individuals. Jason Hong, Anind Dey, Matthew Kam, Aniket Kittur, Robert
Kraut, Jennifer Mankoff and Haakon Faste: Thank you.
I got through the ups and downs of the Ph.D. program only thanks to my
marvelous friends and colleagues at CMU. Tawanna Dillahunt and I shared many
laughs and anguishes. Brian Lim took pictures of our cohort, creating permanent
reminders of many cherished memories. Bryan Pendleton always made sure our
cohort did not starve like graduate students, ate healthily, and slept well. Thank
you.
Scott Davidoff introduced me to American dual-income families’ lives, New York,
American Sci-Fi novels, Citizen Cope, and the charms of Pittsburgh. Turadg
Aleahmad, Gahgene Gweon, Gary Hsieh, Soojin Jun, SeungJun Kim, Sunyoung
Kim, Matthew Lee, Johnny Lee, Joonhwan Lee, Ian Li, Bilge Mutlu, Peter Scupelli,
and Karen Tang have always been quick to offer a hand, and were all good friends
and mentors. Thank you.
Jennifer Marlow, Ruogu Kang, Chloe Fan, Rebecca Gulotta, Anthony Chen,
Gabriela Marcu, and Sauvik Das are energetic individuals who made my daily life
in the HCI Institute colorful. Julia Schwartz, Suyoun Kim, and Stephen Oney
helped me keep going strong with writing routines. Queenie Kravitz created an
environment where all students, including myself, could find home at work.
David Casillas, Kevin Topoloski, and Mark Penney made sure everything went
smoothly. Thank you.
While our interactions were brief, I still remember kind words and advice from
these genuine individuals: Cristen Torrey, Justin Weisz, Andrew Ko, Moira Burke,
4
Amy Hurst, Aruna Balakrishnan, Jeff Wong, and Elijah Mayfield. Their words
were timely, and it kept me going at critical moments. Thank you.
Miso Kim and I shared many long Pittsburgh days and nights working together
and talking about a wide variety of topics. Esther Ahn and I helped each other
stay awake for many sleepless nights as master’s students. Wooyoung Lee, Jueun
Lee, Sooho Park, Takyeon Lee and Suyoun Kim helped me maintain a healthy
body and mind. Thank you.
I had incredible opportunities to work with and learn from many research
assistants: Stephanie Brown, Pong Sarun Savetsila, John Antanitis, Kimberley
Nederlof, Andy Echenique, Sean Kim, Jane Park, Yiwen Jia, Leonard Turnier,
Katherine Cuti, Mingwei Hsu, Mitchell Luban, Alex McCluskey, Ahmad
Shamsuddin, Junior Baboolall, Victoria Yew, and the Snackbot development
team. Thank you.
My final thanks go to my family – my parents, my brother, and Junsung Kim.
5
Abstract
Personalization technology has the potential to optimize service for each person’s
unique needs and characteristics. One way to optimize service is to allow people
to customize the service themselves; another is to proactively tailor services based
on information provided by people or inferred from their past behaviors. These
approaches function best when people know what they want and need, and when
their behaviors and preferences remain consistent over context and time.
However, people do not always know what they want or need, and their
preferences often change. In addition, people cannot always articulate their
preferences with the level of detail required for customization. The customized
service that they want may be suboptimal for their needs. Finally, personalized
services may become obsolete as people’s preferences or contexts change, unless
systems can detect these changes.
This thesis recasts personalization technology to accommodate uncertainties and
changes in people’s preferences and goals. I study personal service providers as a
model for adaptive personalization that helps people customize their services and
that adjusts service according to changes in people’s preferences and goals. I
derive design strategies for adaptive personalization, two of which I empirically
evaluate.
The first strategy adapts service interaction styles to support long-term service
usage. The first two studies investigate ways to detect people’s preferred
interaction styles with a robotic service – whether people treat the system as a
relational being or a utilitarian tool – and the efficacy of personalizing service
interaction based on this interaction preference. The next study explores how the
relational interactions of technology service should be personalized over time in
the context of a robotic snack delivery service in a workplace. Two types of
adaptive relational interaction are investigated in a longitudinal field experiment
6
– a social interaction strategy that adapts its conversation topic to knowledge
common to an organization, and a personalized interaction strategy that learns
about people over time and adapts its interactions accordingly. The results
suggest that social and personalized strategies collectively improve people’s
cooperation, rapport, and engagement with the service over time; the strategies
also influenced social dynamics in the workplace, facilitating the adoption of a
robot into an organization.
The second strategy helps people gain insight into their needs and goals when
they personalize service offerings. This strategy promotes reflection, helping
people think through and articulate their needs and goals. I investigate different
design variables for implementing a reflective strategy for technology service. I
empirically evaluate its efficacy in the context of Fitbit, a physical activity
monitoring service.
This thesis makes contributions to HCI, HRI, and interaction and service design.
It broadens the concept of personalization discussed in HCI and HRI; designs
and evaluates adaptive personalization strategies that accommodate uncertainties
and changes in people’s preferences; draws attention to the dynamic nature of
people’s orientations to interactive technologies; and captures the human-
centered design process of creating and implementing a robotic service.
7
Table of Contents
1 Introduction 14
1.1 Thesis Problem and Approach 16
1.2 Thesis Contributions 19
1.3 Thesis Outline 21
2 Service Approach to Personalization 24
2.1 Personalization Research 24
2.1.1 What Is Personalization 24
2.1.2 Personalization Technology: User-Driven and System-
Driven Personalization 29
2.2 Service Research 33
2.2.1 What Is Service 33
2.2.2 Technology-Based Service and Service Design Research 34
2.2.3 How a Service Approach Can Help 37
3 Adaptive Service Design 40
3.1 Understanding Service Experiences as Dynamic 40
3.1.1 Service Orientation 40
3.1.2 User Experience Over Time 43
3.2 Designing Adaptive Technology-Based Service 46
3.2.1 Line of Adaptivity 46
3.2.2 A Blueprint for A Robotic Snack Delivery Service 47
4 Detecting Service Orientation with Technology-Based
Service 53
4.1 Service Provided by Autonomous Agents 54
4.2 Service Orientation and Interaction Scripts 55
8
4.2.1 Roboceptionist Scripts 56
4.2.2 Greeting as an Indicator of the Script 58
4.2.3 Hypotheses 59
4.3 Method 59
4.3.1 Roboceptionist 60
4.3.2 Data Collection and Coding 61
4.4 Results 63
4.4.1 Grounding Behavior 64
4.4.2 Relational Behavior 66
4.4.3 Conversation Topics 69
4.4.4 Sentence Structure 71
4.5 Discussion and Limitations 71
4.6 Design Implications 73
4.7 Summary 75
5 Matching Interaction Style to Service Orientation 76
5.1 Research Context: Service Breakdowns and Recovery 76
5.2 Mitigation Strategies 78
5.2.1 Expectancy-Setting Strategies 79
5.2.2 Recovery Strategies 79
5.2.3 Service Orientation 81
5.3 Study Design 81
5.3.1 Participants 82
5.3.2 Materials 82
5.3.3 Scenarios 83
5.3.4 Procedure 85
5.3.5 Measures 85
5.4 Results 87
9
5.4.1 Evaluation of the Robots 87
5.4.2 Impact of Service Breakdown 88
5.4.3 Impact of Expectancy-Setting (Forewarning) Strategy 89
5.4.4 Impact of Recovery Strategies 90
5.4.5 Service Orientation and Recovery 91
5.5 Discussion 93
5.6 Implications 94
5.7 Summary 95
6 Snackbot: Design of Robotic Platform and Service for Long-
Term Interaction 96
6.1 Design Approaches to Robotic Systems 96
6.2 Context of Use 98
6.3 Design Goals 99
6.4 Snackbot Design Team 100
6.5 System Overview 101
6.5.1 Hardware 101
6.5.2 Software 102
6.5.3 Form 102
6.5.4 Interaction 103
6.6 Design Process 103
6.6.1 Needs Analysis and Service Concept Generation 105
6.6.2 Observation of Hot Dog Vendor’s Interactions with
Customers 105
6.6.3 Form Giving and Interaction Design 111
6.6.4 Second Prototype 119
6.7 Lessons Learned 123
10