Published by Atlantis Press, a publishing company established in 2006, with offices in Amsterdam, Beijing, and Paris, this book is written in English, but it would certainly benefit from a more thorough editorial contribution by a native English speaker. Myself not being a native English speaker, I humbly acknowledge with gratitude the generosity of most of those who are, with which they have been tolerating my grave errors. But, on the other hand, we who are not should keep on trying to make the distance between the language we use and a standard English (whatever this may mean) as close as possible.
The book is intended to give a systematic overview of text visualization techniques. To accomplish this in just 114 pages is a kind of challenge (the book appears as the first volume in the series “Atlantis Briefs in Artificial Intelligence”). The authors (Cao is with the College of Design and Innovation at TongJi University in China, and Cui is with Microsoft Research Asia in Beijing) coped with it, producing a very up-to-date and information-rich survey, which is indeed brief. They structured their opus into six chapters. In the introduction, they position text visualization into a wider context of information visualization. They also give a book outline, which unfortunately is factually incorrect. It mentions seven chapters, but obviously the concluding one has not been included in the final version (where was the editor, if there was one?). The structure of the remaining content appears to be designed in two levels of detail. In chapter 2, an overview of text visualization techniques is presented in some 30 pages. In this overview, they concentrate specifically on techniques for visualizing document similarity, revealing text content, visualizing sentiments and emotions, and document exploration. Their presentation is brief but very informative. Their choice of groups of visualization techniques largely reflects the intensity of current research in the area.
The remaining four chapters elaborate some of the groups of techniques presented in the overview in more detail. However, first, in chapter 3, a data model is discussed that would be useful to represent text intended for visualization. This is a major research issue because it involves converting unstructured documents into a structured form. Moreover, ultimately the semantic relationships among keywords, sentences, higher structures formed from sentences such as paragraphs or chapters, and among whole documents need to be identified but usually remain implicit. Devoting just nine pages to this issue may seem insufficient, but having in mind the brevity of the whole book, it is acceptable.
Chapters 4, 5, and 6 are devoted to visualizing document similarity, visualizing document content, and visualizing sentiments and emotions. Because the themes have been already presented within the overview in chapter 2, the only possible justification for their inclusion should be in elaborating them in more detail. Therefore, it is very surprising and unfortunate to find not a more detailed elaboration, but a verbatim copy of parts of chapter 2 (in particular, of subsection 2.4) in chapter 6 (in subsection 6.1). Again, what could be a plausible editorial decision to repeat some two pages, including identical figures, in the text twice? Alas, this and other editorial imperfections are completely unnecessary.
Despite some editorial imperfections, the book fulfills its declared aim. It is recommended to anyone who seeks a very informed and up-to-date overview of text visualization techniques.