[empire 2016] adapt to emotional reactions in context-aware personalization
TRANSCRIPT
![Page 1: [EMPIRE 2016] Adapt to Emotional Reactions In Context-aware Personalization](https://reader033.vdocument.in/reader033/viewer/2022042801/589bf5911a28ab40348b6215/html5/thumbnails/1.jpg)
Adapt to Emotional Reactions In Context-aware Personalization
Yong ZhengIllinois Institute of Technology
Chicago, IL, USA
The 4th Workshop on Emotions and Personality in Personalized Systems (EMPIRE), September 16, 2016
![Page 2: [EMPIRE 2016] Adapt to Emotional Reactions In Context-aware Personalization](https://reader033.vdocument.in/reader033/viewer/2022042801/589bf5911a28ab40348b6215/html5/thumbnails/2.jpg)
Emotions
2
![Page 3: [EMPIRE 2016] Adapt to Emotional Reactions In Context-aware Personalization](https://reader033.vdocument.in/reader033/viewer/2022042801/589bf5911a28ab40348b6215/html5/thumbnails/3.jpg)
Emotional Reactions
3
![Page 4: [EMPIRE 2016] Adapt to Emotional Reactions In Context-aware Personalization](https://reader033.vdocument.in/reader033/viewer/2022042801/589bf5911a28ab40348b6215/html5/thumbnails/4.jpg)
Summary
4
We use emotions as context in recommender systems
• Recommender Systems
• Context-aware Recommender Systems
• What is Context?
Adapt to emotional reactions
• Emotional reactions
• The LDOS-CoMoDa Data
• Predictive Models Utilizing Emotional Reactions
• Findings and Results
![Page 5: [EMPIRE 2016] Adapt to Emotional Reactions In Context-aware Personalization](https://reader033.vdocument.in/reader033/viewer/2022042801/589bf5911a28ab40348b6215/html5/thumbnails/5.jpg)
Summary
5
We use emotions as context in recommender systems
• Recommender Systems
• Context-aware Recommender Systems
• What is Context?
Adapt to emotional reactions
• Emotional reactions
• The LDOS-CoMoDa Data
• Predictive Models Utilizing Emotional Reactions
• Findings and Results
![Page 6: [EMPIRE 2016] Adapt to Emotional Reactions In Context-aware Personalization](https://reader033.vdocument.in/reader033/viewer/2022042801/589bf5911a28ab40348b6215/html5/thumbnails/6.jpg)
Recommender Systems (RecSys)
6
RecSys: Item Recommendations to the end users
![Page 7: [EMPIRE 2016] Adapt to Emotional Reactions In Context-aware Personalization](https://reader033.vdocument.in/reader033/viewer/2022042801/589bf5911a28ab40348b6215/html5/thumbnails/7.jpg)
How it works
7
Binary FeedbackRatings Reviews Behaviors
• User Preferences
Explicit Implicit
![Page 8: [EMPIRE 2016] Adapt to Emotional Reactions In Context-aware Personalization](https://reader033.vdocument.in/reader033/viewer/2022042801/589bf5911a28ab40348b6215/html5/thumbnails/8.jpg)
Non-context vs Context
8
Companion
User’s decision may vary from contexts to contexts
• Examples: Travel destination: in winter vs in summer
Movie watching: with children vs with partner
Restaurant: quick lunch vs business dinner
Music: for workout vs for study
![Page 9: [EMPIRE 2016] Adapt to Emotional Reactions In Context-aware Personalization](https://reader033.vdocument.in/reader033/viewer/2022042801/589bf5911a28ab40348b6215/html5/thumbnails/9.jpg)
What is Context?
9
• “Context is any information that can be used to characterize the situation of an entity” by Anind K. Dey, 2001
• Observed Context:
Contexts are those variables which may change when a same
activity is performed again and again.
![Page 10: [EMPIRE 2016] Adapt to Emotional Reactions In Context-aware Personalization](https://reader033.vdocument.in/reader033/viewer/2022042801/589bf5911a28ab40348b6215/html5/thumbnails/10.jpg)
What is Context?
10
Activity Structure:
1). Subjects: group of users
2). Objects: group of items/users
3). Actions: the interactions within the activities
Which variables could be context?
1). Attributes of the actions
Watching a movie: time, location, companion
Listening to a music: time, occasions, etc
2). Dynamic attributes or status from the subjects
User emotions
Yong Zheng. "A Revisit to The Identification of Contexts in Recommender Systems", IUI 2015
![Page 11: [EMPIRE 2016] Adapt to Emotional Reactions In Context-aware Personalization](https://reader033.vdocument.in/reader033/viewer/2022042801/589bf5911a28ab40348b6215/html5/thumbnails/11.jpg)
Emotion in RecSys
11
1). Emotions are helpful in personalization2). Emotions can be considered as effective contexts in RecSys
![Page 12: [EMPIRE 2016] Adapt to Emotional Reactions In Context-aware Personalization](https://reader033.vdocument.in/reader033/viewer/2022042801/589bf5911a28ab40348b6215/html5/thumbnails/12.jpg)
Summary
12
We use emotions as context in recommender systems
• Recommender Systems
• Context-aware Recommender Systems
• What is Context?
Adapt to emotional reactions
• Emotional reactions
• The LDOS-CoMoDa Data
• Predictive Models Utilizing Emotional Reactions
• Findings and Results
![Page 13: [EMPIRE 2016] Adapt to Emotional Reactions In Context-aware Personalization](https://reader033.vdocument.in/reader033/viewer/2022042801/589bf5911a28ab40348b6215/html5/thumbnails/13.jpg)
Emotions and Emotional Reactions
13
![Page 14: [EMPIRE 2016] Adapt to Emotional Reactions In Context-aware Personalization](https://reader033.vdocument.in/reader033/viewer/2022042801/589bf5911a28ab40348b6215/html5/thumbnails/14.jpg)
Emotions in User Interactions
14
Tkalcic, Marko, Andrej Kosir, and Jurij Tasic. "Affective recommender systems: the role of emotions in recommender systems." Proc. The RecSys 2011 Workshop on Human Decision Making in Recommender Systems. 2011.
![Page 15: [EMPIRE 2016] Adapt to Emotional Reactions In Context-aware Personalization](https://reader033.vdocument.in/reader033/viewer/2022042801/589bf5911a28ab40348b6215/html5/thumbnails/15.jpg)
Emotions in User Interactions
15
Entry: before movie watching
Consumption: during movie watching
Exit: after movie watching, e.g., user post-ratings
![Page 16: [EMPIRE 2016] Adapt to Emotional Reactions In Context-aware Personalization](https://reader033.vdocument.in/reader033/viewer/2022042801/589bf5911a28ab40348b6215/html5/thumbnails/16.jpg)
Emotional Expression and Reactions
16
![Page 17: [EMPIRE 2016] Adapt to Emotional Reactions In Context-aware Personalization](https://reader033.vdocument.in/reader033/viewer/2022042801/589bf5911a28ab40348b6215/html5/thumbnails/17.jpg)
Emotional Expression and Reactions
17
User may have similar rating behaviors but different emotional expressions or reactions: WHY??????????
• Different Expectations and Outcomes
– Happy: Well, it is good movie!
– Sad: A sad story. I was moved by the movie
– Surprised: Better than I thought…
• Different User Personality in Emotional Expressions H. S. Friedman and S. Booth-Kewley. Personality, type a behavior, and coronary heart disease: the
role of emotional expression. Journal of Personality and Social Psychology, 53(4):783, 1987. L. Harker and D. Keltner. Expressions of positive emotion in women’s college yearbook pictures and
their relationship to personality and life outcomes across adulthood. Journal of personality and social psychology, 80(1):112, 2001
![Page 18: [EMPIRE 2016] Adapt to Emotional Reactions In Context-aware Personalization](https://reader033.vdocument.in/reader033/viewer/2022042801/589bf5911a28ab40348b6215/html5/thumbnails/18.jpg)
The LDOS-CoMoDa Movie Rating Data Set
18
LDOS-CoMoDa data set, http://www.ldos.si/comoda.html
![Page 19: [EMPIRE 2016] Adapt to Emotional Reactions In Context-aware Personalization](https://reader033.vdocument.in/reader033/viewer/2022042801/589bf5911a28ab40348b6215/html5/thumbnails/19.jpg)
The LDOS-CoMoDa Movie Rating Data Set
19
Pre-Emotions
• Mood
Post-Emotions
• domEmo emotional state during the process
• endEmo final emotional state after the process
Emotional Reactions
• It’s defined as emotions in response to items or user activities, i.e., expression of post-emotions
![Page 20: [EMPIRE 2016] Adapt to Emotional Reactions In Context-aware Personalization](https://reader033.vdocument.in/reader033/viewer/2022042801/589bf5911a28ab40348b6215/html5/thumbnails/20.jpg)
Utilize Emotional Reactions in CARS
20
Emotional Reactions
• Emotions in response to items or user activities
Assumptions:
• Users may have different (even reversed) emotional reactions, but they may have similar rating behaviors finally
• For example, user’s post-emotions may be negative, but finally still leave positive ratings
![Page 21: [EMPIRE 2016] Adapt to Emotional Reactions In Context-aware Personalization](https://reader033.vdocument.in/reader033/viewer/2022042801/589bf5911a28ab40348b6215/html5/thumbnails/21.jpg)
Assumption Validation by Data
21
Post-emotion: negative + rating: positivePost-emotion: positive + rating: negative
Unusual case
![Page 22: [EMPIRE 2016] Adapt to Emotional Reactions In Context-aware Personalization](https://reader033.vdocument.in/reader033/viewer/2022042801/589bf5911a28ab40348b6215/html5/thumbnails/22.jpg)
Utilize Emotional Reactions in CARS Algorithm
22
Emotional Users
![Page 23: [EMPIRE 2016] Adapt to Emotional Reactions In Context-aware Personalization](https://reader033.vdocument.in/reader033/viewer/2022042801/589bf5911a28ab40348b6215/html5/thumbnails/23.jpg)
Utilize Emotional Reactions in CARS Algorithm
23
Model-1: emotional regularization
• Emotional User = User + Post-Emotion
• Post-Emotion could be either domEmo or endEmo
• We measure the similarities between emo users
• We assume user’s rating deviation in corresponding post-emotional state should be similar, if two emotional users are similar
Context-aware Matrix Factorization
![Page 24: [EMPIRE 2016] Adapt to Emotional Reactions In Context-aware Personalization](https://reader033.vdocument.in/reader033/viewer/2022042801/589bf5911a28ab40348b6215/html5/thumbnails/24.jpg)
Utilize Emotional Reactions in CARS Algorithm
24
Model-1: emotional regularization
• We assume user’s rating deviation in corresponding post-emotional state should be similar, if two emotional users are similar
as weight
![Page 25: [EMPIRE 2016] Adapt to Emotional Reactions In Context-aware Personalization](https://reader033.vdocument.in/reader033/viewer/2022042801/589bf5911a28ab40348b6215/html5/thumbnails/25.jpg)
Utilize Emotional Reactions in CARS Algorithm
25
Model-2: emotion + user regularization
• In addition to similar emotional users, the original users may be similar to some extent
as weight
![Page 26: [EMPIRE 2016] Adapt to Emotional Reactions In Context-aware Personalization](https://reader033.vdocument.in/reader033/viewer/2022042801/589bf5911a28ab40348b6215/html5/thumbnails/26.jpg)
Results and Findings
26
domEmo_B: model with emotional regularization onlydomEmo_B,u: model with emotional and user regularizations
We chose domEmo and endEmo as post-emotion respectively
Findings1) There are improvements
2) domEmo is more effective
![Page 27: [EMPIRE 2016] Adapt to Emotional Reactions In Context-aware Personalization](https://reader033.vdocument.in/reader033/viewer/2022042801/589bf5911a28ab40348b6215/html5/thumbnails/27.jpg)
Conclusions
27
• Assumptions: Users may have different (even reversed) emotional reactions, but they may have similar rating behaviors finally
• We validate this assumption in LDOS-CoMoDa data and utilize it to build emotional regularization in context-aware matrix factorization algorithms
• We demonstrate the improvements and discover domEmo is more effective than endEmo
![Page 28: [EMPIRE 2016] Adapt to Emotional Reactions In Context-aware Personalization](https://reader033.vdocument.in/reader033/viewer/2022042801/589bf5911a28ab40348b6215/html5/thumbnails/28.jpg)
Future Work
28
• Emotional Transitions?
– In this work, we just focus on the emotional expressions at the exit stage
– We did not take pre-emotions into account
• By taken pre-emotions into account
– We may build finer-grained models
– We can better find similar emotional users
– But we also have to deal with sparsity problems
![Page 29: [EMPIRE 2016] Adapt to Emotional Reactions In Context-aware Personalization](https://reader033.vdocument.in/reader033/viewer/2022042801/589bf5911a28ab40348b6215/html5/thumbnails/29.jpg)
Adapt to Emotional Reactions In Context-aware Personalization
Yong ZhengIllinois Institute of Technology
Chicago, IL, USA
The 4th Workshop on Emotions and Personality in Personalized Systems (EMPIRE), September 16, 2016