Workshop collocated with ECAI 2012, August 27-31, Montpellier, France
Planlearn serves as a forum for research on intelligent algorithm selection and construction of data mining workflows. It brings together the scientific communities of machine learning, meta-learning and planning, as well as industrial partners aiming to provide intelligent data mining advice to end users.
PlanLearn invites both junior and senior researchers, from both academia and industry, covering research as well as applications of machine learning, to come together and discuss their current work in an interactive setting.
In the 2012 edition, we focus on fostering interaction. The plenary program will feature three invited talks on key issues of current research, as well as short tutorials to provide novel participants an overview of the state-of-the art in the field, and demos of exciting new systems currently under development. In addition, a selection of the best full papers contributed will be eligible for plenary presentation, while other current work in the area will be presented in a poster session. All accepted papers, including those presented as a poster, will be published in the workshop proceedings, and appear in the CEUR Workshop Proceedings. At the end of the workshop, we will hold a panel discussion to chart the most fruitful research directions and establish new collaborations.
PlanLearn 2012 is a full-day workshop on August 28th. It will take place at the University of Montpellier's Campus 2, Building #16, Room TD16.50.
| 9:00 | Workshop Opening |
| 9:05 | Keynote Presentation
'Opportunities for automated workflow planning in the Robot Scientist' by Ross King (University of Manchester) |
| 10:05 |
'Formal Frame for Data Mining with Association Rules - a Tool for Workflow Planning'
by Jan Rauch and Milan Simunek |
| 10:30 | Coffee Break |
| 11:00 |
'Combining Meta-Learning and Optimization Algorithms for Parameter Selection' by T. Gomes, P. Miranda, R. Prudencio, C. Soares, A. Carvalho |
| 11:25 |
'Predicting the Accuracy of Regression Models in the Retail Industry' by Fabio Pinto, Carlos Soares |
| 11:50 |
'Selecting Classification Algorithms with Active Testing on Similar Datasets' by Rui Leite, Pavel Brazdil, Joaquin Vanschoren |
| 12:15 | Lunch |
| 14:00 | Keynote Presentation 'Planning to learn: Recent developments and future directions.' by Filip Zelezny (Czech Technical University) |
| 15:00 |
'Designing KDD-Workflows via HTN-Planning for Intelligent Discovery Assistance' by Jorg-Uwe Kietz, Floarea Serban, Abraham Bernstein, Simon Fischer |
| 15:25 |
'Experimental Evaluation of the e-LICO Meta-Miner' by Phong Nguyen, Alexandros Kalousis, Melanie Hilario |
| 15:50 | Coffee Break |
| 16:10 |
'The Experiment Database for Machine Learning' by Joaquin Vanschoren |
| 16:35 | Panel Discussion |
| 17:00 | Closing |
| 18:00 | ECAI Opening Session |
| 20:00 | Welcome Cocktail Buffet |
Motivation and Topic. Many data mining problems of today require that solutions be elaborated in the form of workflows which include many different processes or operations. Constructing such workflows requires extensive expertise, and could be greatly facilitated by leveraging planning and machine learning techniques. In a data mining workflow, the stage of model building is often preceded by various stages of data preprocessing. An important issue is how these multiple learning processes and operations can be put together. Each task is resolved using certain orderings of operations. The area of planning provides a principled method that can be exploited in this process. Meta-learning can be useful in this process too. It can help to retrieve previous solutions conceived in the past and to re-use them in new settings. This task is inherently interdisciplinary, as it builds on expertise in various areas of AI. Besides the areas mentioned, the area of intelligent design and automatic programming are also relevant.
Aim. The aim of the workshop is to explore the possibilities of this relatively new area, offer a forum for exchanging ideas and experience concerning the state-of-the-art, permit to bring in knowledge gathered in different but related and relevant areas and outline new directions for research. It is expected that the workshop will help to strengthen the sub-communities of researchers interested in exploring these new venues and thus give rise to novel kinds of machine learning/ data mining systems which will provide end users with clear guidance on which techniques to use.
Submissions are possible either as a full paper or as an extended abstract. Full papers should present original, unpublished work, covering research or a case application. Extended abstracts may present current, recently published or future research, and can cover a wider scope. For instance, they may be position statements, offer a specific scientific or business problem to be solved by machine learning / data mining or describe a machine learning / data mining demo or art installation.
A selection will be made of the best paper and runner ups, and these will be presented in the plenary session. The remainder of accepted submissions will be presented in the form of short announcement talks and a poster session. All accepted papers, including those presented as a poster, will be published in the workshop proceedings, and appear in the CEUR Workshop Proceedings, ISSN 1613-0073. The papers selected for plenary presentation will be specially marked in the proceedings.
Full papers can consist of a maximum of 8 pages, extended abstracts up to 2 pages, in the ECAI format. Each submission must be submitted online via the Easychair submission interface. Submissions can be updated at will before the submission deadline. Electronic versions of accepted submissions will also be made publicly available on the conference web site. The only accepted format for submitted papers is PDF. Each paper submission will be evaluated on the basis of relevance, significance of contribution, technical quality, scholarship, and quality of presentation, by at least two members of the program committee. All accepted submissions will be included in the conference proceedings. At least one author of each accepted full paper or extended abstract is required to attend the workshop to present the contribution.
Other areas may be covered, provided they are relevant towards the overall aims of the workshop.
Don't hesitate to contact us!