Book titles for the natural and technical sciences do not include information about all topics a book contains. For example, for a book entitled XML Technology: Principles and Applications in Practice, it is unclear whether a chapter on language XQuery is included or not. Even bibliographic records with subject headings cannot solve such problems because catalogers often do not have sufficient knowledge of particular scientific disciplines. Unlike titles, book tables of contents (ToCs) in the natural and technical sciences often have highly precise descriptions of topics covered in chapters and subchapters. This means it is possible to successfully harvest keywords from ToCs with high relevancy. Such keywords can then become reliable input data for subject indexing and subject classification in library catalogs and other search engines, giving users the ability to search all important topics covered anywhere in a book. Library collections of print books are traditionally processed for library catalogs in the form of bibliographic records consisting of fields with descriptive data (author name, number of pages, language of publication, name of publisher) and subject data (keywords, subject headings, classification terms). Selected fields are indexed and can serve as access points or filters/facets in a catalog. Users are usually looking for either a specific book in a catalog (e.g., a user looks for a book entitled Nanotechnology in biology and medicine: methods, devices, and applications by Tauna Vo-Dinh, 2007) or is looking for books on a certain topic (e.g., a user wants some books in the field of nanotechnology in medicine) [Lown & Sierra & Tito, 2013]. These are two completely different search strategies. In this article, we will look at a technology that addresses the second strategy, a search based on book subject.