Diversity Analysis of Web Search Results

Kingrani, Suneel Kumar; Levene, Mark; Zhang, Dell

doi:10.1145/2786451.2786502

Cited by 6 publications

(4 citation statements)

References 2 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Indeed, the idea of diversified rankings with respect to the preferences of a population appears in several literatures. For example, in the context of search engines, this aim is often referred to as diversifying search results (Welch, Cho, and Olston, 2011;Santos, MacDonald, and Ounis, 2015;Kingrani, Levene, and Zhang, 2015;Wang, Luo, and Yu, 2016). There are also models which incorporate this idea into online advertising (see the work of Hu et al, 2011, and the references therein).…”

Section: Introductionmentioning

confidence: 99%

Proportional Rankings

Skowron

Lackner

Brill

et al. 2017

Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence

View full text Add to dashboard Cite

We extend the principle of proportional representation to rankings: given approval preferences, we aim to generate aggregate rankings so that cohesive groups of voters are represented proportionally in each initial segment of the ranking. Such rankings are desirable in situations where initial segments of different lengths may be relevant, e.g., in recommender systems, for hiring decisions, or for the presentation of competing proposals on a liquid democracy platform. We define what it means for rankings to be proportional, provide bounds for well-known aggregation rules, and experimentally evaluate the performance of these rules.

show abstract

Section: Introductionmentioning

confidence: 99%

Proportional Rankings

Skowron

Lackner

Brill

et al. 2017

Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence

View full text Add to dashboard Cite

show abstract

“…In recent years, a variety of quantitative measures of diversity have been successfully applied in computer science for web search, [20][21][22][23][24] text mining, [25] and recommender systems. [26] Although there exist many different diversity measures (such as Simpson's and Shannon's) and it is debatable which diversity index is the best, [27,28] we choose to use Rao's quadratic entropy [29] to measure the diversity of data, because it takes into account both the sizes of species (groups) and the distances between species (groups).…”

Section: Rao's Quadratic Entropymentioning

confidence: 99%

Estimating the number of clusters using diversity

Kingrani

Levene

Zhang

2017

AIR

Self Cite

View full text Add to dashboard Cite

It is an important and challenging problem in unsupervised learning to estimate the number of clusters in a dataset. Knowing the number of clusters is a prerequisite for many commonly used clustering algorithms such as k-means. In this paper, we propose a novel diversity based approach to this problem. Specifically, we show that the difference between the global diversity of clusters and the sum of each cluster's local diversity of their members can be used as an effective indicator of the optimality of the number of clusters, where the diversity is measured by Rao's quadratic entropy. A notable advantage of our proposed method is that it encourages balanced clustering by taking into account both the sizes of clusters and the distances between clusters. In other words, it is less prone to very small "outlier" clusters than existing methods. Our extensive experiments on both synthetic and real-world datasets (with known ground-truth clustering) have demonstrated that our proposed method is robust for clusters of different sizes, variances, and shapes, and it is more accurate than existing methods (including elbow, Caliński-Harabasz, silhouette, and gap-statistic) in terms of finding out the optimal number of clusters.

show abstract

“…They found that the results returned by Web search engines for rare queries are not as valuable for tail queries and that users issuing tail queries tend to reformulate their queries more often. In a comparative study of the diversity of Web search results in Google and Bing, Kingrani et al (2015) investigated how the two search engines provide diversity in their search results for various categories of Web queries. They found that Google provides more diverse results in the top 50 search results, whereas Bing is more diverse in top 10 search results.…”

Section: Related Researchmentioning

confidence: 99%

The long tail of search and topical queries in public libraries

Oliphant

Shiri

2017

View full text Add to dashboard Cite

Purpose The purpose of this paper is to elaborate the long tail of topical search queries, including the influence of current events, posed to a large, urban public library discovery system. Design/methodology/approach Search queries from the months of June, July, August and September 2014 (1,488,339 total queries) were collected from the Edmonton Public Library’s BiblioCommons database using Google Analytics and exported to Excel. The data were then analyzed using descriptive statistics, frequency counts and textual analysis to explicate the long tail of search, (including the most popular searches) and to explore the relationship between topical search queries and current events. Findings The findings support the long tail theory, as the aggregate tail of topical search queries comprised the vast majority of the total searches and current events exert some influence on the nature and frequency of topical searches. Research limitations/implications Data collection was limited to four months of the year; thus, comparisons across the year cannot be made. There are practical implications for public libraries in terms of marketing and collections, as well as for improving catalogue functionality, to support user search behaviour. Originality/value Not much research attention has been focused on the nature of topical search queries in public libraries compared to academic libraries and the Web. The findings contribute to developing insight into the divergent interests of intergenerational public library users and the topics of materials they are searching for.

show abstract

Diversity Analysis of Web Search Results

Cited by 6 publications

References 2 publications

Proportional Rankings

Proportional Rankings

Estimating the number of clusters using diversity

The long tail of search and topical queries in public libraries

Contact Info

Product

Resources

About