Dear Colleagues,

Springer Publishers and I are considering book proposal submissions for monographs or contributed volumes for the recently launched Book Series entitled "Unsupervised and Semi-Supervised Learning." The series website is http://www.springer.com/series/15892

Below are some interesting facts about Springer and a brief description of the series.

Springer is the largest scientific, technical, and medical publisher in the world. SpringerLink is one of the leading science portals that includes 10+ million documents, an ebook collection with 215,000+ titles, journal archives digitized back to the first issues in the 1840s, 42,000+ protocols and 600+ reference works.

Labeling training data for supervised learning can be expensive, difficult, tedious, error-prone, and even dangerous. With the proliferation of massive amounts of unlabeled data in many application domains, unsupervised learning algorithms that can automatically discover interesting and useful patterns in such data have gained popularity among researchers and practitioners. These algorithms have found numerous applications including pattern recognition, market basket analysis, web mining, social network analysis, information retrieval, recommender systems, market research, intrusion detection, and fraud detection. The difficulty of developing theoretically sound approaches that are amenable to objective evaluation has resulted in the proposal of numerous unsupervised learning algorithms over the past half-century.

Unsupervised algorithms achieve learning without reference to class labels. Semi-supervised algorithms, on the other hand, can make use of both labeled and unlabeled data. This can be useful in application domains where unlabeled data is abundant, yet it is possible to obtain only a small amount of labeled data. Compared to supervised and unsupervised learning, semi-supervised learning is a relatively unexplored subfield of machine learning.

The goal of Springer’s Unsupervised and Semi-Supervised Learning book series is to cover the latest theoretical and practical developments in unsupervised and semi-supervised learning. The intended audience includes students, researchers, and practitioners.

Topics of interest in include:

- Unsupervised/Semi-Supervised Discretization
- Unsupervised/Semi-Supervised Feature Extraction
- Unsupervised/Semi-Supervised Feature Selection
- Association Rule Learning
- Semi-Supervised Classification
- Semi-Supervised Regression
- Unsupervised/Semi-Supervised Clustering
- Unsupervised/Semi-Supervised Anomaly/Novelty/Outlier Detection
- Evaluation of Unsupervised/Semi-Supervised Learning Algorithms
- Applications of Unsupervised/Semi-Supervised Learning

Note that while the series focuses on unsupervised and semi-supervised learning, outstanding contributions in the field of supervised learning will be considered as well.

If you are interested in submitting a proposal for consideration for the series, please let me know and I will email a proposal form to you. If you have any questions, please do not hesitate to contact me.

Thank you for your consideration and I hope to hear from you soon.

M. Emre Celebi, Ph.D.
Series Editor
Professor and Chair
Department of Computer Science
University of Central Arkansas