Discourse Analytics
Carolyn Penstein RoséLanguage Technologies Institute
and Human-Computer Interaction Institute
1
2
Developing technology capable of shaping conversation and supporting effective participation in conversation to achieve positive impact on…
Human learning
Health
Wellbeing
3
AutomaticAnalysis
Of Conversation
ConversationalInterventions
PositiveLearningOutcomes
Computer Supported Collaborative
Learning
** Students learn up to 1.25 standard deviations more when interactive support is provided in the environment. (more than a full letter grade!)
7 Years of Positive Results• Foundational study: students work with a partner and
dialogue agent for support Learn 1.24 s.d. more than individuals without support (Kumar et al., 2007a)
• Selected results inform iterative design of agent behavioro Personalized agents increase supportiveness and help exchange
between students (Kumar et al., 2007b)o Agents are more effective when students have control over timing of
the interaction (Chaudhuri et al., 2008; Chaudhuri et al., 2009)o Agents that employ Balesian social strategies are more effective than
those that do not (Kumar et al., 2010; Ai et al., 2010)o Students are sensitive to agent rhetorical strategies such as displayed bias
(Ai et al., 2010), displayed openness to alternative perspectives (Kumar et al., 2011), and targeted elicitation (Howley et al., 2012)
• Bazaar architecture enables efficient, principle based agent development (Kumar & Rosé, 2011; Adamson & Rosé, 2012)
6
Sociolinguistics
Discourse Analysis
LanguageAnd
Identity
LanguageUse
MachineLearning
Multi-Level
Modeling
AppliedStatistics
ComputationalModels
OfDiscourseAnalysis
Authority
Authority
Engagement Engagement
SouFLé Framework (Howley et al., 2013)
7
Transactive Knowledge Integration
Person Person
8
Easy UI for:• Feature Extraction• Model Building• Error Analysis• Data Structuring
Free/open-source for easy integration with existing code
Modular for fast prototyping and plugin writing in Java
www.lightsidelabs.com
9
Thanks!