SamSut19: Sampling Techniques for Supervised or Unsupervised Tasks |
Submission link | https://easychair.org/conferences/?conf=samsut19 |
Abstract registration deadline | March 31, 2018 |
Submission deadline | July 31, 2018 |
This is the site for the book entitled: "Sampling Techniques for Supervised or Unsupervised Tasks".
With the proliferation of massive amounts of unlabeled data, algorithms that can automatically discover interesting and useful patterns in such data have gained popularity among researchers and practitioners. There is a growing demand for efficient and scalable solutions in many data application domains like biology, computer vision, astronomy or social networking among others. While it is still possible to optimize and speed up some existing techniques, sampling appears as an interesting alternative to manage large data sets. It can be seen as a preprocessing step for supervised and unsupervised algorithms, and as a brick of a more general framework.
Defining a sample that behaves like the whole data set is a quite long-standing issue in data management. There is a real challenge for finding fast, relevant and user-friendly processing techniques for managing large databases. This book will reflect all these recent developments in a comprehensive way and discuss sampling in its different dimensions with a machine learning orientation. It should be accessible to practitioners and useful to anyone teaching or learning pattern recognition or interested in the big data challenge.
The goal of this volume is then to summarize the state-of-the-art in sampling for unsupervised and supervised tasks. Topics of interest include, but are not limited to:
- Overview or survey: sampling techniques for learning algorithms
- Strategies and sampling techniques for unsupervised cases (clustering): distance/density concepts, novel stratification algorithms, new heuristics to find the critical sampling
- Sampling and unsupervised feature selection
- Scalability and curse of dimensionality for unsupervised cases: simultaneous selection of prototype and features for data visualization, clustering...
- Strategies and sampling techniques for supervised cases: simultaneous prototype and feature selection for classification, modelling (Speed up machine learning algorithms such as Nearest classifiers, Naïve-Bayes, Sparse Discriminant analysis, ANNs, regression models...)
- Sampling algorithms for complex data (graphs, objects…): metrics, data fusion, data integration systems, distributed systems, computing architectures
- Sampling algorithms for data streams
- Applications in various fields (internet traffic, smart phone, marketing, medicine...)
Submission Guidelines
The guidelines for contributors to books by Springer are available at:
https://www.springer.com/us/authors-editors/book-authors-editors/book-manuscript-guidelines
Editors
- Frédéric Ros: https://www.researchgate.net/profile/Frederic_Ros
- Serge Guillaume: http://ser.gui.free.fr/homepage