This collection of papers describes current work in the cyber security area. It contains 12 academic papers, with detailed mathematical explanations of the algorithms offered. It is recommended reading for its narrow target audience, an academic community with prior knowledge of the esoterica of machine learning and cyber warfare. A wider audience would find that it assumes too much prior knowledge; the poor grammar can also be distracting.

Machine learning applied to cyber security is a relatively new field. It was discussed in detail in a 2009 book [1], and was also covered at the 2013 IEEE International Conference on Technologies for Homeland Security (HST).

This work suffers from a lack of attention to detail in editing, specifically inadequate proofreading. The eighth paper by Kumar and Kumar is notable in this context. It has instances of overly long sentences, with clauses that differ in voice and tense. Number disagreement between subject and verb is frequent enough to be distracting to the reader. There are also occurrences of correctly spelled but still incorrect words.

The eleventh paper, by Holsopple et al., on the other hand, is an excellent example of what a paper should be. Its English is impeccable, and important terms are defined early in the paper to aid understanding.

At less than 300 pages, the book is of sufficient length to cover the material presented. The academic convention of referencing is followed throughout. Illustrations are sufficient in number, generally clear, and relevant to the context. There is no index.

In summary, the 12 papers are mixed in quality. Some introduce new methods, while some discuss existing work. Some show evidence of thorough proofreading, while some do not. If you have a background and interest in the subject matter, then this book may be a worthwhile addition to your bookshelf.