Jihie Kim's Home Page

I am the Principal Investigator of the Pedagogical Technologies (PedTek) group in the USC Information Sciences Institute (ISI). I am also a Research Assistant Professor in the Computer Science Department at the University of Southern California (USC).

I received my Ph.D. in Computer Science from University of Southern California. I received Masters and Bachelor's degrees in Computer Science and Statistics from the Seoul National University .

I am organizing the 2nd workshop on Intelligent Support for Learning in Groups at AI in Education Conference (AIED 2013).

Recent Papers on Social Dialogue and Collaboration

Current research focuses on Social Dialogue and Collaboration. Research areas covered include online discourse analysis, forum and wiki mining, mixed-initiative interfaces, workflow systems, and interactive knowledge capture.

This paper describes instructional tools that help students and instructors use discussion boards more effectively, with an emphasis on automatically assessing discussion activities, and promoting student discussion participation and interaction by connecting information seekers to information providers.

• Jihie Kim and Erin Shaw, Pedagogical Discourse: Connecting Students to Past Discussions and Peer Mentors within an Online Discussion Board, The 21st Innovative Applications of Artificial Intelligence Conference (IAAI-2009), 2009.

This paper describes a novel topic modeling approach where we induce topic categories from a canonical text of the given domain and use it to develop discussion topic classifiers with little training data.

• Donghui Feng, Jihie Kim, Erin Shaw, and Ed Hovy, Towards Modeling Threaded Discussions through Ontology-based Analysis, In Proceedings of National Conference on Artificial Intelligence (AAAI-2006) , 2006.

This paper describes typical emotion acts (EAs) that exhibit in student online discussions and illustrates how emotion acts can be used in assessing and predicting student discussion outcome.

• Jihie Kim, Erin Shaw, Saul Wyner, Taehwan Kim, and Jia Li, Discerning Affect in Student Discussions, Annual Meeting of the Cognitive Science Society, 2010.

This paper presents a new assessment framework for e-learning where we facilitate robust assessment of student learning through workflow framework.

• Jun Ma, Erin Shaw and Jihie Kim, Computational Workflows for Assessing Student Learning, The 10th International Conference on Intelligent Tutoring Systems Conference, 2010.

This paper presents a new knowledge acquistion dialogue framework where the system incorporates instructional strategies that good learners take in proactively reasoning about knowledge acquisition tasks and dynamically presenting suggestions. It provides a good example of how we can bring research results from teaching and learning back to AI research.

• Jihie Kim and Yolanda Gil, Incorporating Tutoring Principles into Interactive Knowledge Acquisition, International Journal of Human-Computer Studies , Volume 65, Issue 10, October 2007, Pages 852-872.

Research Areas

The following describes my work in individual areas.

Online Forum/Wiki mining and Collaboration analysis:

• Scaffolding pedagogical discourse in engineering courses. Assessing student activities in threaded on-line discussions and providing answers to student queries. See more details in these papers:
  • Analyzing Answers in Threaded Discussions using a Role-Based Information Network (SocialCom 2011).
  • Discerning Affect in Student Discussions (CogSci 2010).
  • Pedagogical Discourse: Connecting Students to Past Discussions and Peer Mentors within an Online Discussion Board (IAAI 2009).
  • Scaffolding On-line Discussions with Past Discussions: An Analysis and Pilot Study of PedaBot, (ITS 2008).
  • Profiling Student Interactions in Threaded Discussions with Speech Act Classifiers, (AIED 2007).
  • An Intelligent Discussion-Bot for Answering Queries in Threaded Discussions (IUI 2006).
  • Learning to Detect Conversation Focus of Threaded Discussions (HLT-NAACL 2006).
  • Towards Modeling Threaded Discussions through Ontology-based Analysis (AAAI 2006).

• Workflows for assessing student learning. Developing a workflow environment that supports efficient assessment of student learning through interactive generation and execution of various assessment workflows.
  • Workflow-Based Assessment of Student Online Activities with Topic and Dialogue Role Classification (AIED 2011).
  • Computational Workflows for Assessing Student Learning (ITS 2010).

Semantic Approaches to Workflows:

• Wings (Workflow INstance Generation and Selection). We present an approach to workflow creation and validation that uses semantic representations to describe complex scientific applications in a data-independent manner, then automatically generates workflows of computations for given data sets, and finally maps them to available computing resources. We have used this to create workflows of thousands of computations, which are submitted to the Pegasus mapping system for execution over grid computing environments. See more details in this paper (IEEE Intelligent Systems article, 2011) and this paper (CCPE article, 2008).

• CAT (Composition Analysis Tool). CAT facilitates interactive construction of workflows (computation pathways) where users select and connect existing workflow components, and the system interactively gernerates assistance in completing a correctly formulated pathway. See more details in this paper(IUI 2004) and this site.

Interactive Knowledge Capture:

• ECHO/ SLICK (Skills for Learning to Interactively Capture Knowledge). Developing acquisition interfaces that are proactive learners, able to reason about learning activities and with initiative in participating in the process accordingly. See more details in this article (IJHCS 2007), this paper(AIEd 2003) and these sites: Slick and Echo.

• KANAL (Knowledge ANALysis). A Tool for checking process models entered by users. By relating different pieces of information in process models among themselves and to the existing KB, it performs a variety of verification and validation checks and propose useful fixes. See more details in this paper (IJCAI 2001) and this site.

• EMeD(Expect Method Developer). A tool to guide knowledge base creation based on interdependency analysis. See more details in this paper(AAAI 1999) and this site.

• KA (Knowledge Acquisition) Evaluation Methodology. See more details in this paper (JETAI article 2001).

Knowledge-Based Approach for Component Modeling:

• Active Catalogs that provides on-line catalogs augmented by behavioral models and their consumption environment, significantly enhancing the engineering and design support that is desirable but beyond the reach of current engineering environments. See more details in this paper (AI EDAM article 2004).

Thesis work:

Utility Problem in Soar (Expensive Chunks). My dissertation research centers on application of machine learning techniques to speed up problem solving. Many learning systems suffer from the utility problem; that is, that time after learning is greater than time before learning. Discovering how to assure that learned knowledge will in fact speed up system performance has been a focus of research in explanation-based learning (EBL). One way of finding a solution which can guarantee such cost boundness is to analyze all the sources of cost increase in the learning process and then eliminate these sources. I began on this task by decomposing the learning process into a sequence of transformations that go from a problem solving episode, through a sequence of intermediate problem solving/rule hybrids, to a learned rule. This transformational analysis itself is important to understand the characteristics of the learning system, including cost changes through learning. Such an analysis has been performed for Soar (a problem solving system with a variant of EBL). By analyzing these transformations, I have identified a set of sources which can make the output rule expensive. Also, I have implemented modifications of the learning system based on the analysis. The work is summarized in this paper (AI Journal article)

Robot research:

• the YODA robot project. A mobile agent that can autonomously learn from its environment based on its own actions, percepts, and missions. (article in USC Chronicle) Also see this paper (AI magazine article).

Grants

• PedEval: Measuring the Impact of 0nline Discourse in Undergraduate STEM Courses: Semi-automatic Assessment of Large Discussion Board Corpora
  Principal Investigator. National Science Foundation, REESE Program [Award #1008747] .
  
  Studying effectiveness of communication as a means for measuring student learning and the further development of novel metrics based on NLP and machine learning techniques.

• PedGames: Creative Secondary STEM Learning Through Collaborative Game Building
  Principal Investigator. National Science Foundation, Creative IT Program [Award #1002901] .
  
  Developing and evaluating school curriculum for learning standards-based content via game making. The project addresses the educational needs of at-risk high school students in underperforming schools in the Los Angeles Unified School District (LAUSD).

• SocRecomm: Mining and Supporting Teacher Social Networking: Intelligent Recommendation of NSDL Resources to support Middle School Teacher Collaboration
  Principal Investigator. National Science Foundation, NSDL Program[Award #1044427].
  

  Realizing the goal of self educating teacher communities by providing intelligent recommendation services that will find the best connections between mentors and mentees, and find the most relevant NSDL resources and forums that meet the needs of individual teachers. Such recommendation systems will greatly enhance the usability and usefulness of the current educational social networks.

• PedWiki: Instructional Assessment of Wiki Use in Engineering Courses through Discourse Analysis and Topic Classification
  Principal Investigator. National Science Foundation, The Course Curriculum and Laboratory Improvement Program (CCLI) program [Award #0941950].
  
  The project focuses on three inter-related ways in which Wiki use can be exploited for education. We propose to do the following: 1. Develop and study new and ongoing Wiki-based engineering projects, and compare their adoption to Wiki projects in non-engineering courses. 2. Create instructional (pedagogical) assessment tools based on discourse analysis and course topic ontology for qualitatively evaluating student Wiki use. 3. Identify scaffolding opportunities such as automated recommender interfaces to promote student engagement and collaboration.

• PedWorkflow: Workflows for assessing student learning.
  Principal Investigator. National Science Foundation, CISE Information and Intelligent Systems (IIS) program [Award #0917328].
  
  Developing a workflow environment that supports efficient assessment of student learning through interactive generation and execution of various assessment workflows.

• PedDiscourse: Scaffolding pedagogical discourse in engineering courses.
  Principal Investigator. National Science Foundation, The Course Curriculum and Laboratory Improvement (CCLI) phase II program [Award #0618859]).
  
  Assessing student activities in threaded on-line discussions and providing answers to student queries. The goal of the project is to design, deploy and evaluate software tools that automatically structure and scaffold undergraduate student interactions within online discussion boards.

Address:
Computer Science Department and Information Sciences Institute
University of Southern California
4676 Admiralty Way
Marina del Rey, CA 90292