Staging Mixed-intiative Dialogs by Program Generation and Transformation

Our research project seeks to improve the design and implementation of human-computer dialogs, such as those used in ATMs, airport kiosks, and smart phones, and especially mixed-initiative dialogs, through the creative, and often non-traditional, use and application of concepts and techniques from programming languages such as currying and partial evaluation. We are optimisitc that the philosophical and conceptual connections between natural and programming languages suggest that additional concepts from programming languages, such as reflection, lazy evaluation, and continuations, will find a natural place in our methodology. We see particular promise in the use of first-class continuations to elegantly model the transfer of control between dialog participants. Our long-term research goal is to develop and evaluate computational models, based on programming languages concepts, to simplify the complexity involved in designing and implementing mixed-initiative human-computer dialogs.

Below you will find links to our dialog toolkit, publications, and the webpages of team members.

Dialog Modeling and Management Toolkit

Our dialog modeling and management toolkit is available here.

The toolkit consists of a dialog miner and stager generator (see figure below). The dialog miner losslessly compresses a high-level specification of a dialog into an a minimal set of partially ordered sets (or posets), i.e., an optimal, intermediate, implementation-neutral, representation of a dialog specification. Given a set of posets, the stager generator produces a stager capable of staging the interaction required to complete the dialog.

Papers published by ACM Press are copyright by the ACM. PDF versions of these papers are posted here by permission of ACM for your personal use. Not for redistribution.

Perugini, S. & Ramakrishnan, N. (2010). Program transformations for information personalization. Accepted for publication in Computer Languages, Systems and Structures. To appear [DOI].

Perugini, S., Anderson, T.J., & Moroney, W.F. (2007). A study of out-of-turn interaction in menu-based, ivr, voicemail systems. Proceedings of the 25th International ACM Conference on Human Factors in Computing Systems, 961-970. New York, NY: ACM Press. [DOI | PDF].

Perugini, S. (2006). Real-time query expansion and procedural interfaces for information hierarchies. Proceedings of the International ACM SIGIR Workshop on Faceted Search, 50-54 [PDF].

Perugini, S. & Ramakrishnan, N. (2005). A generative programming approach to interactive information retrieval: Insights and experiences. Proceedings of the 4th International ACM Conference on Generative Programming and Component Engineering, LNCS 3676, 205-220. Berlin: Springer. [DOI | PDF].

Perugini, S. & Ramakrishnan, N. (2005). Personalization by Program Slicing. Journal of Object Technology, 4(3), 5-11. (Special issue on the 6th International ACM GPCE Young Researchers Workshop, Vancouver, Canada.) [PDF | HTML].

Narayan, M., Williams, C., Perugini, S., & Ramakrishnan, N. (2004). Staging transformations for multimodal web interaction management. Proceedings of the 13th International ACM World Wide Web Conference, 212-223. New York, NY: ACM Press. [DOI | PDF].

Perugini, S. McDevitt, K., Richardson, R., Pérez-Quiñones, M.A., Shen, R., Ramakrishnan, N., Williams, C., & Fox, E.A. (2004). Enhancing usability in CITIDEL: Multimodal, multilingual, and interactive visualization interfaces. Proceedings of the 4th International ACM/IEEE-CS Joint Conference on Digital Libraries, 315-324. New York, NY: ACM Press. [DOI | PDF].

Perugini, S. & Ramakrishnan, N. (2003). Personalizing web sites with mixed-initiative interaction. IEEE IT Professional, 5(2), 9-15. (Featured on the cover of the March-April issue and recognized in ACM TechNews Timely Topics, 5(490), Friday, May 2, 2003.) [DOI | PDF].

Capra, R., Narayan, M., Perugini, S. Ramakrishnan, N., & Pérez-Quiñones, M.A. (2003). The Staging Transformation Approach to Mixing Initiative. Working Notes of the IJCAI Conference, Workshop on Mixed-Initiative Intelligent Systems, 23-29 [PDF].

Faculty

Saverio Perugini (Assistant Professor, Univ. of Dayton, Dept. of Computer Science)
Naren Ramakrishnan (Professor, Virginia Tech, Dept. of Computer Science)
William F. Moroney (Emeritus Associate Professor, Univ. of Dayton, Dept. of Psychology)
Manuel A. Pérez-Quiñones (Associate Professor, Virginia Tech, Dept. of Computer Science)

Students

John-Paul V. Cresencia (M.C.S, Univ. of Dayton, Dept. of Computer Science, expected May 2010)
Michael Narayan (Ph.D. student, Virginia Tech, Dept. of Computer Science)

Former student members

Shuangyang Yang (M.S., Univ. of Dayton, Dept. of Electro-optics, Dec 2008)
Now a Ph.D. student in the Holcombe Dept. of Electrical and Computer Engineering at Clemson University.
Taylor J. Anderson (MA, Univ. of Dayton, Dept. of Psychology, September 2006)
Thesis: A Study of Out-of-turn Interaction with a Voicemail System using Speech Recognition
Now employed at SA Technologies in Marietta, GA.
Robert G. Capra III (Ph.D., Virginia Tech, Dept. of Computer Science, February 2006)
Thesis: An Investigation of Finding and Refinding Information on the Web
Now post-doctoral fellow in Gary Marchionini's group in the School of Information and Library Science at the Univ. of North Carloina Chapel Hill.
Chris Williams (M.S., Virginia Tech, Dept. of Computer Science, May 2004)
Thesis: WS://IM: A Software Framework for Multimodal Web Interaction Management
Now at Cryptek, Inc. in Sterling, VA.
Saverio Perugini (Ph.D., Virginia Tech, Dept. of Computer Science, May 2004)
Thesis: Program Transformations for Information Personalization
Now Assistant Professor at Univ. of Dayton, Dept. of Computer Science.
Atul Shenoy (M.S., Virginia Tech, Dept. of Computer Science, June 2003)
Thesis: A Software Framework for Out-of-turn Interaction in a Multimodal Web Interface
Now at Microsoft, Inc. in Redmond, WA.