Measuring Diversity in Heterogeneous Information Networks

by   Pedro Ramaciotti Morales, et al.

Diversity is a concept relevant to numerous domains of research as diverse as ecology, information theory, and economics, to cite a few. It is a notion that is continuously gaining attention in the information retrieval, network analysis, and artificial neural networks communities. While the use of diversity measures in network-structured data finds a growing number of applications, no clear and comprehensive description is available for the different ways in which diversities can be measured in data described by these structures. In this article, we develop a formal framework for the application of a large family of diversity measures to heterogeneous information networks (HINs), a flexible, widely used, network data formalism. This allows for an extension of the application of diversity measures, from systems of classifications and apportionments, to more complex relations that can be better modeled by networks. In doing so, we do not only provide an effective organization of multiple practices from different domains, but we also unearth new observables in systems modeled by heterogeneous information networks. The pertinence of the approach is illustrated by the development of different applications related to various domains concerned by both diversity and networks. In particular, we illustrate the usefulness of these new proposed observables in the domains of recommender systems and social media studies among other fields.


page 1

page 2

page 3

page 4


Testing the Impact of Semantics and Structure on Recommendation Accuracy and Diversity

The Heterogeneous Information Network (HIN) formalism is very flexible a...

Fairness and Diversity in Recommender Systems: A Survey

Recommender systems are effective tools for mitigating information overl...

Recommenders with a mission: assessing diversity in newsrecommendations

News recommenders help users to find relevant online content and have th...

Modeling the Evolution of Networks as Shrinking Structural Diversity

This article reviews and evaluates models of network evolution based on ...

Representation Learning for Recommender Systems with Application to the Scientific Literature

The scientific literature is a large information network linking various...

Digital Forensics Domain and Metamodeling Development Approaches

Metamodeling is used as a general technique for integrating and defining...

Next Priority Concept: A new and generic algorithm computing concepts from complex and heterogeneous data

In this article, we present a new data type agnostic algorithm calculati...

Please sign up or login with your details

Forgot password? Click here to reset