Privacy-Enhanced Public Name-Authority System for Building Research Communities

Abstract

Today, the Internet has become an important source of information about academic researchers and their research activities.  The types of information one can obtain from the Internet include contact information, publication information, biographical information, photographs, and other miscellaneous information. Some of this information is generated by professional societies and academic institutions, while other information is generated by individuals and independent enterprises.

As the quantity of academic material on the web grows, finding and processing information about a researcher's work is increasingly difficult. For example, it is often hard to discern whether authors of different papers in tangentially-related areas are the same person, based solely on a name. Even if one can determine this information, it is often difficult to assess the accuracy of information obtained, especially if it was generated either by an individual or by community of users.

In this thesis, we propose a novel community-based and web-accessible repository of information about academic researchers and their research activities. First, we introduce a web-application called \emph{Federated World Directory of Mathematicians}(FWDM), which retrieves personal information from a variety of disparate data-sets, and which inspired the solutions proposed in this thesis. We then propose a \emph{public name-authority system}, as a means to provide high quality disambiguated information on researchers. The proposed system helps to ensure the quality of information by obtaining only the information approved by the research community. We introduce and describe two approaches to the design of public name-authority systems - the data-filtered and the user-filtered name authority systems - in order to explore their benefits and drawbacks.