DIGITAL LIBRARY
COMPASS: A WEB-BASED ANCILLARY APPLICATION FOR COLLEGE TEXTBOOKS
University of Patras (GREECE)
About this paper:
Appears in: INTED2019 Proceedings
Publication year: 2019
Pages: 578-587
ISBN: 978-84-09-08619-1
ISSN: 2340-1079
doi: 10.21125/inted.2019.0223
Conference name: 13th International Technology, Education and Development Conference
Dates: 11-13 March, 2019
Location: Valencia, Spain
Abstract:
Students in higher education institutes usually have to study textbooks of hundreds of pages. Moreover, in the case of particular disciplines - like for example, history of art - companions contain a large number of images which accompany and further explain the text. In this setting, a difficulty experienced by the majority of students consists in searching and grouping image items based on specific features. For example, it is not at all obvious how one can fast and easily locate and list all paintings created by a particular artist in a 1000 page textbook containing 500 illustrations. A straight-forward approach could be exhaustive manual search, which would be highly inefficient though. Index tables and glossaries can certainly be helpful; however, they can only support very limited search patterns, like for example, “find pages where a term appears”.

Motivated by a similar difficulty experienced by students of our department regarding the book “The Story of Art” by E.H. Gombrich [G95] used as the main companion in the context of an Art History course, we designed and developed Compass, a companion e-assistant for large college textbooks. It is essentially a web-based application built on html5, php and sql which can assist the study of a large textbook supporting a search facility for image elements. More precisely, Compass currently runs on our department web server. Extensive information on images included in the textbook is maintained in the Compass database. The application can be accessed through a simple, mobile-friendly interface. Users can search the collection by submitting keywords regarding artist name, type of creation (e.g., painting, sculpture, etc), estimated creation date, artistic movement, short description, etc. The application returns a list of results matching the search keywords. Furthermore, results are ordered and presented in groups based on the position of keywords in the full description of the items.

Currently, Compass can indicatively support search in the Greek and English version of the book “The Story of Art” by E.H. Gombrich. A demo can be found at: http://web4.culture.uwg.gr/compass/. However, the application can further support search in multiple textbooks as long as detailed information about their image content is loaded in a corresponding database. Towards this aim, we have also developed a web-based auxiliary tool for creating and populating a Compass database.

So far, the evaluation of Compass shows that it can certainly make a useful ancillary application for students. We plan to enrich the database for the particular textbook and extend Compass in order to support search in additional textbooks. We also plan to extend the evaluation of our application including people with additional roles (like for example, teachers, researchers, etc). To the best of our knowledge no similar application has been suggested so far. We believe that web-based search applications for large college textbooks containing extensive image content could significantly facilitate and support study and research.

References:
[G95] E. H. Gombrich. The Story of Art. Phaidon Press, 1995, 688 pp.
Keywords:
Web-based application, html5, php, sql, college textbooks, art, database, search, semantics, image collections.