ABSTRACT
This final Chapter extends some of the techniques developed throughout the book for schemas and mappings to a new kind of digital artifacts, namely digital identities. In the following, we tackle the problem of disambiguating identifiable entities on the Web. A user-driven scheme is proposed where graphs of identities self-organize in a dynamic and probabilistic manner. The solution has two desirable properties:
(i) it allows end-users to freely define associations between arbitrary identities, and
(ii) it probabilistically infers identity relationships based on uncertain links using constraint-satisfaction mechanisms.
