The thesis begins by exploring entity-entity association, and by considering entity networks as a means of exploring document collections, and formulate the task as ranking related entities. The task of impact-based entity recommendation is formalized and two approaches are proposed based on learning to rank and impact propagation.
The second theme concerns entity-document associations. This type of association is studied in the context of filtering documents for knowledge base acceleration. In this setting, the goal is to filter documents that are relevant to update a profile of an entity. The focus is on the challenge of long-tail entities specifically, and propose an approach that leverages intrinsic, i.e. in-document signals more.
Finally, entity-aspect associations are explored. Entities are often associated with attributes, types, distinguishing features, topics, or themes. This type of information is grouped under the heading ‘aspect’.’ Entity aspects are analysed in the context of Web search, and defined as common search tasks in the context of an entity.
This thesis contributes new task formalizations, algorithms, and insights on computing entity associations for search. The experimental results confirm the effectiveness of the approaches within different settings. Insights gained from this thesis will help address entity-oriented information access challenges in various domains.