|
|
 |
My Research and Collaborations
Qualitative Modeling from Scattered Data |
Collaborators: Daniel Lemire (UQAM),
Martin Books (NRC)
Visions:
Qualitative models are more suitable than classical quantita-tive models in many tasks like Model-based Diagnosis(MBD), explain-ing system behavior, and designing novel devices from first principles. Monotonicity is an important feature to leverage when constructing qual-itative models. Detecting monotone pieces robustly and efficiently from sensor or simulation data remains an open problem. This paper intro-duces an approach based on scale-dependent monotonicity: the notion that monotonicity can be defined relative to a scale. Real-valued func-tions defined on a finite
set of reals e.g. the sensor data the simulation results, can be partitioned into quasi-monotone segments, i.e. segments monotone with respect to nonzero scale. We can identify the extrema of the quasi-monotone segments. This method is used to abstract qualitative models from simulation models for the purpose of diagnosis. It shows that using monotone analysis, the abstracted qualita-tive model is not only sound, but also parsimonious because it generates few landmarks.
Expected Results:
the monotonicity by scale theory
the best segmentation algorithm
qualitative modeling from simulation data for diagnosis
the soundness of the qualitative model abstraction
Published:
Martin Brooks, Yuhong Yan, Daniel Lemire, "Scale-Based Monotonicity Analysis in Qualitative Modelling with Flat Segments",IJCAI05,Edinburgh, Scotland, 30 July - 5 August, 2005.(accepted)
Yan, Y., Lemire, D., and Brooks, M. “Monotonicity Analysis for Constructing Qualitative Models”, 2005 Conference on
Model-based Reasoning, Dec. 16-18, 2005, Pavia, Italy (Accepted)
Yuhong Yan, "Sensor Placement and Diagnosability Analysis at Design Stage", MONET Workshop on Model-Based Systems at ECAI 2004, August 22-26, Valencia, Spain.
Yuhong Yan, Daniel Lemire and Martin Brooks "Monotone Pieces Analysis for
Qualitative Modeling", MONET Workshop on Model-Based Systems
at ECAI 2004, August 22-26, Valencia,Spain.
Yuhong Yan, “Qualitative
Model Abstraction for Diagnosis”, the 17th Int.
Workshop on Qualitative Reasoning, August 20-22, 2003, Brasilia, Brazil.
Yuhong Yan, “Automatic Qualitative Model
Abstraction from Numeric Simulation Model”, the 14th Int.
Workshop on Principles of Diagnosis, June 11-14, 2003, Washington DC, USA.
|
Web Services for On-line Experiment Systems
Collaborators: Hamadou Saliah (Teluq)
Students: Xinge Du, Yong Liang
Visions:
An online experiment system provides web-based GUI for the students to operate the remote devices and shares the computational and experimental resources among the online laboratories. Therefore the interoperability and accessibility are two main concerns. Web Service is the emerging technology for this application. Web Service is based on Internet protocols and XML technologies which are platform-independent and language-independent. Web Service can be the middleware for client/server or peer-to-peer architectures.
In this thesis, we will use Web Services to build a online experiment so that different online laboratories can share their experimental resources. We will design a framework for online experiment systems; study the methodology to wrap the instrument functions as Web Services; and benchmark the quality of service (QoS) of Web Services in this kind of application.
This project is a part of the collaborative project of Using Web Services for Online Experiments between NRC-ITT and CLIEF research center in Teluq University.
Expected Results:
The framework of the online experiment system using Web Services
The methodlogy to wrap the instrument functions as Web Services
The matric of QoS of Web Services for online experiments
The benchmark of QoS
Published:
Yuhong Yan, Yong Liang, Xinge Du, "Controlling Remote Instruments Using Web Services for Online Experiment Systems", 3rd International Conference on Web Service, Orlando, Florida, USA, July 11-15, 2005 (accepted).
Yuhong Yan, Xinge DU, Yong Liang, "Distributed and Collaborative Environment for Online Experiment System Using Web Services", CSCWD05, May 24-26, 2005, Coventry, UK (accepted).
Yuhong Yan, Yong Liang, Xinge Du, Hamadou Saliah-Hassane, Ali Ghorbani, "Design Instrumental Web Services for Online Experiment Systems", Ed-Media 2005, Montreal,Canada, June 27-July 2, 2005 (accepted).
Hamadou Saliah-Hassane, Djamal Benslimane, Ileana De La Teja, B. Fattouh, Liên Kim Do, Paquette Gilbert, Maarouf Saad, Louis Villardier, Yuhong Yan, “A General Framework for Web Services and Grid-Based Technologies for Online Laboratories”, iNEER Conference for Engineering Education and Research, March, 2005, Tainan, Taiwan. (Invited Paper)
Techical Report
Magadalena Piasecka, “Online Experiments and Learning Management Systems”.
|
On-line Music Learning Portal
Collaborators: Richard Hornsby (UNB music director), Naxos (Music Publisher)
Students: Matthias Klein, Ning Wang, Han Liang
Visions:
The On-line Music Learning Portal primary objective is to use the multimedia capabilities of the World Wide Web with a mix of off-the-shelf and proprietary software to create a complete musical learning experience for a wide range of people seeking to expand their performance abilities or to beginning music players that not necessarily have access to a direct person-to-person musical education. The target audience for this educational tool is the young and adult audiences with different access to Internet connections, from dial-up to broadband, with an interest in furthering their musical knowledge and performance. The On-line Music Learning Portal will enhance the learning experience with the possibility, in a second phase of the project, for the learner to connect with the teacher through videoconferencing in real-time classes.
Expected Results:
The open platform of on-line music learning portal
Business model to deliver music services and music pieces on-line
Multimedia file management and retrieval
Collaborative environment in music teaching and learning
Published:
Matthias Klein, Yuhong Yan, Ning Wang, Liang Han, "The design of an Online Music Learning Portal", Ed-Media 2005, Montreal, Canada, June 27-July 2, 2005, (accepted).
|
Web services monitoring and diagnosis
Collaborators: Yannick Pencole (the Australian National University)
|
Research Topics for Students
Contact me for the topics.
Here are some advices for graduate students.
|
Contact Info
My current address is:
NRC-IIT
46 Dieneen Drive, Fredericton, NB E3B 9W4, Canada
Phone: +1-(506)-444-0388
Fax: +1-(506) 444-6114
e-mail: yuhong.yan(at)nrc.gc.ca
| |
| |
