Result diversification has recently attracted much attention as a means of increasing user satisfaction in recommender systems and web search. Many different approaches have been proposed in the related literature for the diversification problem. In this paper, we survey, classify and comparatively study the various definitions, algorithms and metrics for result diversification.

Big data technology offers unprecedented opportunities to society as a whole and also to its individual members. At the same time, this technology poses significant risks to those it overlooks. In this article, we give an overview of recent technical work on diversity, particularly in selection tasks, discuss connections between diversity and fairness, and identify promising directions for future work that will position diversity as an important component of a data-responsible society. We argue that diversity should come to the forefront of our discourse, for reasons that are both ethical-to mitigate the risks of exclusion-and utilitarian, to enable more powerful, accurate, and engaging data analysis and use.

Result diversification has recently attracted considerable attention as a means of increasing user satisfaction in recommender systems, as well as in web and database search. In this paper, we focus on the problem of selecting the k-most diverse items from a result set. Whereas previous research has mainly considered the static version of the problem, in this paper, we exploit the dynamic case in which the result set changes over time, as for example, in the case of notification services. We define the CONTINUOUS k-DIVERSITY PROBLEM along with appropriate constraints that enforce continuity requirements on the diversified results. Our proposed approach is based on cover trees and supports dynamic item insertion and deletion. The diversification problem is in general NP-complete; we provide theoretical bounds that characterize the quality of our solution based on cover trees with respect to the optimal solution. Finally, we report experimental results concerning the efficiency and effectiveness of our approach on a variety of real and synthetic datasets.

No abstract

In publish/subscribe systems, users describe their interests via subscriptions and are notified whenever new interesting events become available. Typically, in such systems, all subscriptions are considered equally important. However, due to the abundance of information, users may receive overwhelming amounts of events. In this paper, we propose using a ranking mechanism based on user preferences, so that only top-ranked events are delivered to each user. Since many times top-ranked events are similar to each other, we also propose increasing the diversity of delivered events. Furthermore, we examine a number of different delivering policies for forwarding ranked events to users, namely a periodic, a sliding-window and a history-based one. We have fully implemented our approach in SIENA, a popular publish/subscribe middleware system, and report experimental results of its deployment.

No abstract

Recently, result diversification has attracted a lot of attention as a means to improve the quality of results retrieved by user queries. In this paper, we propose a new, intuitive definition of diversity called DisC diversity. A DisC diverse subset of a query result contains objects such that each object in the result is represented by a similar object in the diverse subset and the objects in the diverse subset are dissimilar to each other. We show that locating a minimum DisC diverse subset is an NP-hard problem and provide heuristics for its approximation. We also propose adapting DisC diverse subsets to a different degree of diversification. We call this operation zooming. We present efficient implementations of our algorithms based on the M-tree, a spatial index structure, and experimentally evaluate their performance.

No abstract