Author ORCID Identifier

http://orcid.org/0000-0001-9139-3433

Year of Publication

2017

Degree Name

Doctor of Philosophy (PhD)

Document Type

Doctoral Dissertation

College

Engineering

Department

Computer Science

First Advisor

Dr. Jerzy W. Jaromczyk

Abstract

Faceted browsing (also called faceted search or faceted navigation) is an exploratory search model where facets assist in the interactive navigation of search results. Facets are attributes that have been assigned to describe resources being explored; a faceted taxonomy is a collection of facets provided by the interface and is often organized as sets, hierarchies, or graphs. Faceted browsing has become ubiquitous with modern digital libraries and online search engines, yet the process is still difficult to abstractly model in a manner that supports the development of interoperable and reusable interfaces. We propose category theory as a theoretical foundation for faceted browsing and demonstrate how the interactive process can be mathematically abstracted in order to support the development of reusable and interoperable faceted systems.

Existing efforts in facet modeling are based upon set theory, formal concept analysis, and light-weight ontologies, but in many regards they are implementations of faceted browsing rather than a specification of the basic, underlying structures and interactions. We will demonstrate that category theory allows us to specify faceted objects and study the relationships and interactions within a faceted browsing system. Resulting implementations can then be constructed through a category-theoretic lens using these models, allowing abstract comparison and communication that naturally support interoperability and reuse.

In this context, reuse and interoperability are at two levels: between discrete systems and within a single system. Our model works at both levels by leveraging category theory as a common language for representation and computation. We will establish facets and faceted taxonomies as categories and will demonstrate how the computational elements of category theory, including products, merges, pushouts, and pullbacks, extend the usefulness of our model. More specifically, we demonstrate that categorical constructions such as the pullback and pushout operations can help organize and reorganize facets; these operations in particular can produce faceted views containing relationships not found in the original source taxonomy. We show how our category-theoretic model of facets relates to database schemas and discuss how this relationship assists in implementing the abstractions presented.

We give examples of interactive interfaces from the biomedical domain to help illustrate how our abstractions relate to real-world requirements while enabling systematic reuse and interoperability. We introduce DELVE (Document ExpLoration and Visualization Engine), our framework for developing interactive visualizations as modular Web-applications in order to assist researchers with exploratory literature search. We show how facets relate to and control visualizations; we give three examples of text visualizations that either contain or interact with facets. We show how each of these visualizations can be represented with our model and demonstrate how our model directly informs implementation.

With our general framework for communicating consistently about facets at a high level of abstraction, we enable the construction of interoperable interfaces and enable the intelligent reuse of both existing and future efforts.

Digital Object Identifier (DOI)

https://doi.org/10.13023/ETD.2017.138

Share

COinS