From Facet-based to Visualised Digital Library Interfaces

Staff - Faculty of Informatics

Date: 3 February 2021 / 14:00 - 16:30

Online

You are cordially invited to attend the PhD Dissertation Defense of Maram Barifah on Wednesday February 3rd, 2021 at 14:00 on MS Teams.

Abstract:
Digital library (DL) systems in the last decade have provided an unprecedented access to library content that were confined to their physical spaces. The last twenty years of the DL system renovation have witnessed several generations of online catalogs from the Online Public Access Catalogue (OPAC) late 1980s, to the next generation catalog systems (NGC). DL systems are extensively used to facilitate access to collections of digital resources. Thus, NGC offers more user-friendly interfaces by providing functionalities like faceted navigation and filters. The facets are based on the unified index of metadata of the library's content, and it is believed that facets assist users to achieve higher task accuracy and satisfaction. However, many researchers criticize the architectural elements of the traditional representation of contents in DL systems for two main reasons: firstly, the knowledge organisation schemes adopted to build DL systems were created by experts to classify resources, this results in inadequate use of the search tools i.e. facets and other DL functions by the end-users who are not familiar with the purpose of such techniques. Secondly, users are not allowed to deploy techniques of berrypicking in the search process, as no other information seeking behaviour (ISB) than the goal-directed ISB are supported. The basic assumption of this study is that there is a need to provide alternative modes of access to information than the traditional facet-based DL interfaces. The context of this study includes three different libraries serving three different communities of users. These libraries have in common similar facet-based interfaces that built on the traditional DL information architecture. Our investigation follows a user-centered approach where the users are a central part in the design process of the system. Therefore, this research employs ethnographic techniques, that aim to explore the interactions of the real users along with other user attributes e.g. experience level, attitude, and affective variables including emotion and mood. In particular, this research investigates how users interact with the DL interfaces, and how they perceive the quality of the DL interface through: log file analysis, screen video recording, user experience (UX) study, and a large attitudinal survey. Results from such investigation lead to the second phase of the thesis which is the redesign of the facet-based interface by adopting the visualization approach. The visualization technique has been selected as it is supposed to provide an intuitive means for users to interact with the information. An intuitive interface is characterized by being easy to use and learn, requiring little to no previous knowledge, and just feeling "natural" to use. Thus, the main concern of this thesis is to examine the effectiveness of the visualized interface as an alternative to the traditional DL interfaces, by considering the users' perspective. The evaluation of the visualized interface indicates that visualization is a promising solution that can enhance positive gauging of the participants' satisfaction, together with experiential values that invoke low level of frustrations and high enjoyment.

Dissertation Committee:
- Prof. Fabio Crestani, Università della Svizzera italiana, Switzerland (Research Advisor)
- Dr. Monica Landoni, Università della Svizzera italiana, Switzerland (Research co-Advisor)
- Prof. Matthias Hauswirth, Università della Svizzera italiana, Switzerland (Internal Member)
- Prof. Vittorio Limongelli, Università della Svizzera italiana, Switzerland (Internal Member)
- Prof. Pia Borlund, Oslo and Akershus University College of Applied Sciences, Norway (External Member)
- Prof. Ian Ruthven, University of Strathclyde, Glasgow, UK (External Member)