posted by user: Bartosz_Krawczyk || 3416 views || tracked by 7 users: [display]

HMLNCD 2013 : Hybrid Machine Learning for Non-stationary and Complex Data

FacebookTwitterLinkedInGoogle

Link: http://dap.vsb.cz/wsc17/
 
When Dec 3, 2012 - Dec 14, 2012
Where on-line
Submission Deadline Nov 7, 2012
Notification Due Nov 19, 2012
Final Version Due Nov 25, 2012
Categories    machine learning   soft computing   pattern recognition   data mining
 

Call For Papers

We invite you tu submit a paper for Hybrid Machine Learning for Non-stationary and Complex Data - a Special Session during the 17th Online World Conference on Soft Computing in Industrial Applications (WSC17). Following the tradition of WSC Conferences Series WSC17 will be carried on-line with the dual benefits of rapid dissemination and no cost from the participants. Papers will be published by Springer within the Advances In Intelligent and Soft Computing series as an edited volume after the conference

The field of machine learning is developing rapidly over the last decade. Achieving at the same time great theoretical advances and numerous practical applications it became one of the most prominent fields of artificial intelligence. Yet there still exists a need for further development as contemporary problems require more sophisticated and carefully designed methods. Hybrid machine learning systems, using techniques from the soft computing area such as evolutionary algorithms or fuzzy logic became one of the most promising research directions in this research area. They allowed to expand the possibilities of canonical learning approaches to deal with imprecision, find optimal solutions or to prune the classifier structures to name a few. Therefore hybrid machine learning methods are an interesting solution to solve contemporary problems connected with the increasing complexity of the data. Many knowledge sources outputs information in a non-stationary patterns such as data streams and time series. Additionally researchers must deal with the ever-increasing complexity and size of the data. This aspects propel the need for compound and dedicated hybrid machine learning algorithms.
This session aims at gathering a collection of top-quality papers written by various groups of researchers. It will bring together new developments in this ever-expanding field and give an overview on the state of the art. This will allow to bring together researchers dealing with similar subjects, to exchange thoughts and to start a lively discussion about the present and future directions in hybrid ML algorithms.

The HMLM Special Session will aim at bringing together a collection of high quality papers dealing with hybrid systems in the analysis of non-stationary and complex data. Papers presenting considerable improvements in the field, as well as practical implementations and real-life evaluations of systems will be considered for publication. We are interested in the fusion of soft computing and machine learning in topics such as e.g. data stream analysis, concept drift, multiple classifier systems, compound pattern recognition, massive data analysis one-class classification or imbalanced data. The topics of interest for this Special Session will be presented below.

The scope of HMLM Special Session covers, but is not limited to, following topics:
A. Hybrid Machine Learning Methods for Non-stationary problems:
- Data stream analysis
- Concept drift detection
- Model selection for streaming data
- Active learning and forgetting for changing environments
- Time series prediction

B. Hybrid Ensemble Methods:
- Ensembles for classification / regression / feature selection etc.
- Classifier selection / diversity / ensemble pruning
- Fusion methods (untrained / trained)
- Bagging / Boosting / Random Forest / Mixture of Experts
- Clustering Ensembles
- Multiple Kernel Methods
- Ensembles of evolutionary algorithms
- Rough set-based ensembles

C. Hybrid one-class classification.

D. Massive data mining
- Dimensionality reduction
- Parallel data analysis
- Large graph structure analysis

E. Real-life problems of non-stationary and complex data analysis form fields such as:
- medicine
- bioinformatics
- chemoinformatics
- sensors
- environmental engineering
- information retrieval
- image analysis

Special Session Organisers:
Bartosz Krawczyk, Wrocław University of Technology, bartosz.krawczyk@pwr.wroc.pl
Michał Woźniak, Wrocław University of Technology, michal.wozniak@pwr.wroc.pl

Important dates:
Submission of papers: November 07, 2012
Notification of acceptance: November 19,2012
Submission of camera-ready papers: November 25,2012
Conference: December 3-14,2012

Paper submission instructions:
Papers will be published by Springer within the Advances In Intelligent and Soft Computing series as an edited volume after the conference.
For details about paper submission refer to the WSC17 web site.
Please upload the papers via the conference submission system.
In case of any questions please contact directly the Special Session Organisers.

Related Resources

ICDMML 2020   【EI SCOPUS】2020 International Conference on Data Mining and Machine Learning
ISBDAI 2020   【Ei Compendex Scopus】2020 International Symposium on Big Data and Artificial Intelligence
ICML 2020   37th International Conference on Machine Learning
ICPR 2020   International Conference on Pattern Recognition 2020
MNLP 2020   4th IEEE Conference on Machine Learning and Natural Language Processing
AICA 2020   O'Reilly AI Conference San Jose
DASFAA 2020   Database Systems for Advanced Applications
MAAIDL 2020   Springer Book 'Malware Analysis using Artificial Intelligence and Deep Learning'
Journal Special Issue 2019   Machine Learning on Scientific Data and Information
IEEE CiSt 2020   6th IEEE Congress on Information Science and Technology