The key to understanding the breathtaking development of mankind lies in the ability to objectify knowledge. Just like human knowledge has undergone changes, so have the libraries in their role as treasure chambers of said knowledge. With the advent of the digital age, traditional libraries were complemented by huge collections of digital documents. These digital libraries, or e-libraries, result from the fact that knowledge can nowadays be objectified in the form of multifaceted digital copies. Relatively speaking, we are still at the dawn of the digital age—let us not forget that the Web is only three decades old—and yet, the documents that have been collected in a digital library display an astonishing range in terms of multimedia feature diversity: traditional texts, photos, and graphics, recorded audio files (e.g., conversations, music, and sounds of all kinds), moving images (e.g. video sequences, films, or live recordings of events), to name but a few. Future developments will certainly lead to further forms and formats; it is therefore safe to say that the portfolio of digital documents will increase exponentially, and so will the number of digital libraries, consequentially.

As a result of this development, modern users are confronted with new challenges that come in the form of explosions of objectified knowledge caused  by digital technology. How does one manage to find the most relevant documentsabout a certain topic? Who is going to assist the user with this procedure? In traditional libraries of the pre-digital era, one could address the librarian and ask for human assistance. The librarian knew the order by whichthe book inventory was arranged and could thus give the required information without being an expert on any of the subject matters. When looking for a similar kind of support when using a digital library, one finds first and foremost high-performance keyword search engines, mostly Google, which, despite their considerable capabilities, exhibit limitations when it comes to interacting with the user; their user interfaces have generally not been optimized for this particular task, and they usually suffer from an inability to put the user’s query into a precise context.

This book introduces a new approach to designing E-Librarian Services. With the help of this system, users will be able to retrieve multimedia resources from digital libraries more efficiently than they would by browsing through an index or by using a simple keyword search. E-Librarian Services combine recent advances in multimedia information retrieval with aspects of human-machine interfaces, such as the ability to ask questions in natural language; they simulate a human librarian by finding and delivering the most relevant documents that offer users potential answers to their queries. The premise is that more pertinent results can be retrieved if the search engine understands the meaning of the query; the returned results are therefore logical consequences of an inference rather than of keyword matches. Moreover, E-Librarian Services always provide users with a solution, even in situations where they are unable to offer a comprehensive answer.