Diversifying web search results

Abstract

Result diversity is a topic of great importance as more facets of queries are discovered and users expect to find their desired facets in the first page of the results. However, the underlying questions of how 'diversity' interplays with 'quality' and when preference should be given to one or both are not well-understood. In this work, we model the problem as expectation maximization and study the challenges of estimating the model parameters and reaching an equilibrium. One model parameter, for example, is correlations between pages which we estimate using textual contents of pages and click data (when available). We conduct experiments on diversifying randomly selected queries from a query log and the queries chosen from the disambiguation topics of Wikipedia. Our algorithm improves upon Google in terms of the diversity of random queries, retrieving 14% to 38% more aspects of queries in top 5, while maintaining a precision very close to Google. On a more selective set of queries that are expected to benefit from diversification, our algorithm improves upon Google in terms of precision and diversity of the results, and significantly outperforms another baseline system for result diversification.

Personalized diversification of search results

“…23 We first investigated creating a set of fixed ambiguous topics using Wikipedia's disambiguation pages, which indicate concepts or topics that may have multiple meanings. This methodology was used in previous evaluation schemes in the literature [14,15]. However, after an initial analysis of the worker's feedback, we found that workers were sometimes unfamiliar with the selected ambiguous topics, and were as a consequence unable to provide consistent relevance judgments.…”

“…For instance, if is an Open Directory Project (ODP) category as in [1,14], we may estimate ( | ) by a text classification method as the probability that document belongs to class . Other estimation approaches can be applied for query aspects of a different nature.…”

“…The implicit assumption is that the loss involved in zero vs. small utility is far more significant than the loss involved between degrees of satisfaction. In other words, relevance follows a diminishing marginal returns pattern, where the gain obtained from increased relevance decreases fast with the number of relevant documents [14,21].…”

Diversifying Product Review Rankings: Getting the Full Picture

“…A greedy algorithm is used for ranking search results such that relevance is maximized while variance is minimized. Rafiei et al [29], introduce a similar framework based on Portfolio Theory for reranking Web search results. The problem of result diversification is also investigated in the area of structured data queries.…”

mNIR: Diversifying Search Results Based on a Mixture of Novelty, Intention and Relevance

“…Finally, they proposed an evaluation methodology to characterize the objectives and the underlying axioms. Rafiei et al [12] modeled the diversity problem as expectation maximization and presented algorithms to estimate the optimization parameters. In [4], documents are selected sequentially according to relevance.…”