News (2007)

News (2007)

MITACS Travel Award - Mateo 2007

Congratulation to Mark Anthony F. Mateo, who won a MITACS Travel Award to attend (1) Minicourse on Quantitative Biology and (2) Workshop on Deconstructing Biochemical Networks both held in Montréal during September 22-28, 2007. Among all graduate students and PDFs attending the minicourse or workshop, Mateo was one of five recipients of this award.

DEXA - KUPA 2007

• 1. Dr. Carson K. Leung serves as a Program Chair for the First International Workshop on Knowledge Management and Discovery for Ubiquitous and Pervasive Applications (KUPA 2007), which is organized in conjunction with the 18th International Conference on Database and Expert Systems Applications (DEXA 2007), held on September 7, 2007 in Regensburg, Germany. The proceedings are published by the IEEE Computer Society.

  2. W05--KUPA'07 (1st International Workshop on Knowledge Management and Discovery for Ubiquitous and Pervasive Applications):

  3. Workshop Introduction and Organizers.

  4. In A M. Tjoa, R.R. Wagner (Eds.):

  5. DEXA Workshops 2007:

  6. 173-176

     1. W05 - KUPA '07

     2. 1st International Workshop on Knowledge Management and Discovery for Ubiquitous and Pervasive Applications

     3. Message from the Program Chairs of KUPA 2007

     4. The First International Workshop on Knowledge Management and Discovery for Ubiquitous and Pervasive Applications (KUPA) is organized in conjunction with the 18th International Conference on Database and Expert Systems Applications (DEXA) in Regensburg, Germany, on September 3-7, 2007.

     5. Due to recent advancements in communication technology as well as in computing and storage resources, ubiquitous and pervasive applications are becoming increasingly popular. The large scale integration and examination of heterogeneous and independent date sources has created new challenges to knowledge management and discovery. The KUPA 2007 workshop brings together academics and industry professionals to discuss recent progress and challenges in knowledge management and discovery in ubiquitous and pervasive environments. This workshop also serves as a platform for theoreticians and practitioners to exchange their ideas, experiences, and opinions on using database and artificial intelligence technologies to manage and discover knowledge for ubiquitous and pervasive applications.

     6. In this workshop, we selected a few papers, which cover a broad range of interesting issues and their progress in very specialized topics about applications of managing and discovering knowledge in ubiquitous and pervasive environments. For example, Baralis et al., in their paper, investigate the issue of reducing energy and bandwidth consumption for query collection in sensor network environments. They exploit clustering techniques to select a good-quality subset of representative sensors from the whole network to reduce communication and computation costs among sensors. Chai et al. propose an algorithm that uses a bipartite graph to find frequent patterns. Their algorithm improves the performance of the frequent-pattern mining process by scanning the transaction database only once and avoiding the generation of candidate patterns. Luo et al. present an improved version of role-based access control model. Qiu et al. show how they use categorization to summarize multiple documents. These papers discuss various techniques for managing knowledge for, and discovering knowledge from, different ubiquitous and pervasive applications.

     7. This workshop would not have been possible without the help and effort of many people. We thank the organization team of the main conference (DEXA 2007) for their support of our workshop. We also express our thanks to the Honorary General Chairs of KUPA 2007 for their valuable advice and suggestions towards this workshop. We are grateful to members of the Program Committee for their professionalism and dedication in the process of judging the contributions of papers and producing constructive comments to the authors.

     8. Sajid Hussain

     9. Acadia University, Canada

     10. Christel Kemke

     11. The University of Manitoba, Canada

     12. Carson K. Leung

     13. The University of Manitoba, Canada

ICMLC 2007

On Wednesday, August 22, 2007, Mark Anthony F. Mateo presented a refereed paper titled "CAMEL: An intelligent computational model for agro-meteorological data", which he co-authored with his academic supervisor (Dr. Carson K. Leung) and their government collaborator (Andrew J. Nadler from MAFRI), in ICMLC 2007 held in Hong Kong, China. The paper was published by the IEEE Press.

MITACS SAC (Aug 2007)

MITACS Student Advisory Committee (SAC) - August 2007

Student Notes

Interview with a MITACS intern

1. Can you please briefly describe your project?

Data mining refers to non­-trivial extraction of implicit, previously unknown, and potentially useful from data. In our project, my academic supervisor (Dr. Carson K. Leung) and I worked with our government partner (Manitoba Agriculture, Food and Rural Initiatives (MAFRI)) to design and develop data mining techniques to agro­-meteorological data provided by MAFRI. Specifically, we came up with mathematical solutions to analyze both current and historic weather data as well as related information. We identified and extracted various explicit, previously unknown, and potentially useful information (such as relationships and trends that exist both spatially and temporally) from the agro-­meteorological data. We also incorporated our mathematical solutions mentioned above; we designed and developed a multi­-algorithm data quality assurance system. The system analyzes and assesses real-­time weather and environmental data and detects and flags weather observations that may be anomalous. In this project, we made use of various databases, data mining, mathematics, and statistics concepts in the development and operation of our system.

2. Who is your industry partner, and why are they interested in this work?

In this project, we worked with Manitoba Agriculture, Food and Rural Initiatives (MAFRI), a provincial government department responsible for the well-being of Manitoban food producers. MAFRI operates a network of weather stations scattered throughout the province of Manitoba. This network provides general monitoring, value-added information, and decision support tools to agricultural producers in Manitoba. The weather data gathered from these stations are used as inputs in MAFRI's agricultural decision models. As the output from these models depends upon the input meteorological observations, more accurate data leads to more accurate model. Uncertainty in the input data will certainly cast doubt on the validity and reliability of the model. To this end, the provincial government department of MAFRI is interested on automated tools to ensure that agro­-meteorological observations fed into the model are clean and error-­free resulting in more accurate and more reliable model.

3. What industrial, social and/or environmental repercussions could your work have?

MAFRI currently operates at least two management practices decision support tools: Potato Disease Severity Model (DSV) and Corn Crop Heat Models (CH). A key objective of our project is to have reliable and error-­free data to improve the accuracy of MAFRI's agricultural decision models.

Acquiring accurate agricultural models allows MAFRI to better administer its DSV and CH advisory programs to Manitoban farmers. Without these models, it is estimated that Manitoba farmers spend $15 million annually for fungicidal programs alone to combat the disease. The DSV helps prevent late blight disease by classifying a risk level which allows farmers to have better timing strategies in their fungicide application. Studies have shown that Manitoban farmers can save up to $2.5 million per year if the potato DSV model is diligently followed.

With more precise prediction tools, MAFRI is in a better position to provide more accurate disease occurrence predictions allowing farmers to improve the management of their fungicide control programs. Having accurate models will also allow MAFRI to investigate which particular geographic region and season could be considered as a 'hotspot' for disease occurrence. In effect, this allows the agency to channel resources more intelligently to a specific time and area in the province where it is genuinely needed. This is indeed very beneficial to Manitoba, especially due to its vast geographical expanse that makes it difficult to have simultaneous farm monitoring. The end result is a better yield for Manitoba farmers effectively transforming into greater profit and benefit for the Manitoba agricultural economy.

Moreover, it is important to note that the system that my academic supervisor and I designed and developed is not confined to just agro­-meteorological data. Our system can certainly be applicable to control data quality in many other practical database and data mining applications (e.g., analysis of utility consumption, monitoring of traffic, and law enforcement in highways).

4. How did you get become involved with the MITACS internship program?

I heard about the MITACS internship program from my academic supervisor (Dr. Carson K. Leung), who initiated this internship collaboration with our partner (MAFRI). As a professor in the Department of Computer Science and a research affiliate in the Institute of Industrial Mathematical Sciences at The University of Manitoba, Dr. Leung was introduced about the internship program by Karen Booth (MITACS representative).

My academic supervisor and I would like to take this opportunity to thank the following people who have helped us during this MITACS internship program: Dr. John A. Bate (Head of CS at UofM), Dr. Abba Gumel (Director of IIMS at UofM), Karen Booth (MITACS), Dr. Laurence Meadows (MITACS), Deanna Lanoway (MITACS), and Mr. Andrew J. Nadler (MAFRI).

5. Have you been involved with MITACS in other ways?

Have you attended any MITACS annual conferences? The system that my academic supervisor and I designed and developed is very effective in controlling data quality. Due to this fruitful result, our partner (MAFRI) nominated us as candidates for Best Use of Mathematics in Technology Transfer Award. We won this MITACS award, which was presented at the 2007 Annual MITACS Conference in Winnipeg.

6. How has the MITACS internship program contributed to your academic development?

Through this MITACS internship, I was able to see how advanced mathematics aid in real-­life situations. Before the internship, my idea of mathematics was only confined to classroom examples and very esoteric theoretical concepts. My internship with MITACS changed this idea as I was able to experience first-­handed how mathematical sciences help organizations such as MAFRI to better achieve their service delivery goals in improving programs for the benefit of their stakeholders. It has allowed me to connect and network with extraordinarily talented and experienced people outside the academe and thus have a hands-­on appreciation on how mathematical tools and technologies are used in real­-life, non­-academic settings.

To summarize, I enjoyed taking part in this MITACS internship program. It is an excellent program to strengthen the connection between my academic supervisor (Dr. Carson K. Leung) and our government partner (MAFRI). We (professor­-partner­-intern) all benefited from this program. This internship program provides my professor (Dr. Leung) an additional opportunity to apply his expertise in advanced level mathematical sciences—­­for example, his expertise in databases and data mining—to areas that address vital research opportunities. It also helps our partner (MAFRI) utilize new mathematical tools and technologies to address advanced scientific issues that are vital to the success of MAFRI and its stakeholders. It as well gave me (as an intern) an opportunity to work on the practical side of my MSc research program. We are happy to recommend the MITACS internship program to our peers. Last but not least, I thank Dr. Leung in giving me the opportunity to work on this interesting research project, to consistently guide me, and to provide me with constructive comments and suggestions throughout this internship. I also thank MITACS and MAFRI for funding this project.

MITACS Award 2007

• 1. Faculty of Science Annual Report: July 1, 2006 to June 30, 2007

     1. XIII SELECTED DEPARTMENTAL HIGHLIGHTS

        • An MSc student in the Department of Computer Science, Mark Mateo, working with Dr. Carson Leung on a MITACS project, won an award for Best Novel Use of Mathematics in Technology Transfer.

  2. President's Report: June 27, 2007

• • 1. II. ACADEMIC MATTERS

    2. Faculty of Science

       • Mark Mateo, a graduate student working with Dr. Carson Leung, Computer Science, has won a 2006 Mathematics of Information Technology and Complex Systems (MITACS) Student Award in the category "Best Novel Use of Mathematics in Technology Transfer".

MITACS Internship 2007

Efficient Mining of Agro-Meteorological Data

Faculty Supervisor: Dr. Carson Leung, University of Manitoba

Intern: Mark Anthony Mateo

Province: Manitoba

University: University of Manitoba

Partner: Manitoba Agriculture, Food and Rural Initiatives (MAFRI)

Type of Internship: Mathematical sciences

Discipline: Computer science

Sector: Information and communications technologies

Program: Accelerate

Data mining refers to non-trivial extraction of implicit, previously unknown, and potentially useful information from data. In this project, the intern will apply data mining techniques to agro-meteorological data provided by Manitoba Agriculture, Food and Rural Initiatives (MAFRI). Specifically, he plans to develop mathematical solutions to analyze both current and historic weather data as well as related information such as crop yields, soil inventories, and crop management practices. These solutions would identify various explicit, previously unknown, and potentially useful information (such as relationships and trends that exist both spatially and temporally) from the agro-meteorological data. We also plan to design, implement, and integrate these techniques into an operational quality assurance and control system whereby real-time weather and environmental data are analyzed and assessed for observations that fall outside the limits imposed by the relationships that have been identified. Then, he will apply this system to assess historic climate data to identify trends, shifts, or patterns and as well as key factors affecting the level of agricultural risks.

Exploration efficace de données agrométéorologiques

Superviseur universitaire: M. Carson Leung, Université du Manitoba

Étudiant/Intern: Mark Anthony Mateo

Province: Manitoba

Université: Université du Manitoba

Partenaire: Agriculture, Alimentation et Initiatives rurales Manitoba (AAIRM)

Discipline: Informatique

Secteur: Agriculture

Programme: Accélération

L'exploration de données se réfère à l'extraction, dans des données, d'information implicite, jusque-là inconnue et potentiellement utile. Dans le cadre de ce projet, le stagiaire appliquera des techniques d'exploration dans des données agrométéorologiques fournies par Agriculture, Alimentation et Initiatives rurales Manitoba (AAIRM). Plus précisément, il prévoit mettre au point des solutions mathématiques permettant d'analyser des données météorologiques actuelles et historiques ainsi que de l'information connexe telle que le rendement des cultures, les inventaires de sols et les pratiques de gestion des cultures. Ces solutions cerneront, dans des données agrométéorologiques, des renseignements précis, jusque-là inconnus et potentiellement utiles (tels que les relations et tendances qui existent sur le plan spatial et temporel). Nous prévoyons également concevoir, mettre en œuvre et intégrer ces techniques dans un système d'assurance et de contrôle de la qualité opérationnelle qui permettra d'analyser et d'évaluer des données météorologiques et environnementales en temps réel pour effectuer des observations à l'extérieur des limites imposées par les relations cernées. Le stagiaire utilisera ensuite ce système pour évaluer des données climatiques historiques en vue de cerner des tendances, des changements ou des modèles ainsi que les facteurs clés qui déterminent les niveaux de risque agricole.

Celebration of Excellence 2007

Congratulation to Mark Mateo, who was recognized as a recipient of the University of Manitoba Graduate Scholarship (UMGS) at the Celebration of Excellence organized by UofM Faculty of Science held Tuesday, February 20, 2007. Among all Master's and PhD students, he was one of nine new recipients of this award.