Federated Search

Shokouhi, Milad; Si, Luo

doi:10.1561/1500000010

Cited by 121 publications

(74 citation statements)

References 204 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We address a particular kind of entity search, namely the search over multiple data sources, called federated search, which entails the three main problems of source representation, source selection, and result merging [8,9]. We focus on the latter for federated entity search in uncooperative settings as illustrated in Figure 1b, where only ranked result lists of entity descriptions are obtained from each source and no further information about the sources is available.…”

Section: Overviewmentioning

confidence: 99%

“…C(e i , e j ) = 1 if r(ei) < r(ej) 0 otherwise (9) The second encoding exploits the difference between ranks to express the degree of how much e i is preferred over e j using discounted ranks (labeled with subscript d as M P M d ):…”

Section: Ranking Evaluationmentioning

confidence: 99%

“…Federated search (distributed IR) has been thoroughly studied for document retrieval [8,9,32], where the unit of retrieval are textual documents and not structured entities. Query translation for federated entity search has been investigated by [30], who use source-specific query generators to adapt a structured query to each source.…”

Section: Related Workmentioning

confidence: 99%

“…This solution and the comprehensive work of the database community in this realm [5][6][7] assume full access to the entire datasets to compute features such as weights of attributes, co-occurences or to learn parameters, which are then used to resolve all coreferences between two or more datasets in one run. However, access to the entire datasets is either not granted in many application scenarios such as search over multiple Web data sources (where data access is only provided via APIs for single requests), also called federated search over uncooperative sources [8,9], or many data sources are highly dynamic, imposing a high burden on batch processing to keep up with frequent changes and to provide fresh information for time sensitive applications such as search over stock quotes, movies and timetables. Distributed document retrieval for uncooperative environments has been studied in the IR community [8,9].…”

Section: Introductionmentioning

confidence: 99%

“…However, access to the entire datasets is either not granted in many application scenarios such as search over multiple Web data sources (where data access is only provided via APIs for single requests), also called federated search over uncooperative sources [8,9], or many data sources are highly dynamic, imposing a high burden on batch processing to keep up with frequent changes and to provide fresh information for time sensitive applications such as search over stock quotes, movies and timetables. Distributed document retrieval for uncooperative environments has been studied in the IR community [8,9]. We investigate the task of federated entity search with structured entities consisting of a varying number of attributes and corresponding values.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Federated Entity Search Using On-the-Fly Consolidation

Herzig

Mika

Blanco

et al. 2013

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. Nowadays, search on the Web goes beyond the retrieval of textual Web sites and increasingly takes advantage of the growing amount of structured data. Of particular interest is entity search, where the units of retrieval are structured entities instead of textual documents. These entities reside in different sources, which may provide only limited information about their content and are therefore called "uncooperative". Further, these sources capture complementary but also redundant information about entities. In this environment of uncooperative data sources, we study the problem of federated entity search, where redundant information about entities is reduced on-the-fly through entity consolidation performed at query time. We propose a novel method for entity consolidation that is based on using language models and completely unsupervised, hence more suitable for this on-the-fly uncooperative setting than state-of-the-art methods that require training data. Further, we apply the same language model technique to deal with the federated search problem of ranking results returned from different sources. Particular novel are the mechanisms we propose to incorporate consolidation results into this ranking. We perform experiments using real Web queries and data sources. Our experiments show that our approach for federated entity search with on-the-fly consolidation improves upon the performance of a state-of-the-art preference aggregation baseline and also benefits from consolidation.

show abstract

Section: Overviewmentioning

confidence: 99%

Section: Ranking Evaluationmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Federated Entity Search Using On-the-Fly Consolidation

Herzig

Mika

Blanco

et al. 2013

Lecture Notes in Computer Science

View full text Add to dashboard Cite

show abstract

To what problem is distributed information retrieval the solution?

Thomas

2012

J Am Soc Inf Sci Tec

View full text Add to dashboard Cite

Distributed information retrieval (DIR), where a single broker coordinates retrieval from many independent search services, has been extensively studied but typically without any particular application and sometimes even without any explicit motivation. There have been a handful of arguments given for DIR-coverage, effectiveness, and ease of use, for example-but these are not borne out by experience.Are there uses for DIR? There are, but generally for organizational not technical reasons, and they have not been well studied.

show abstract

Explicit search result diversification using score and rank aggregation methods

Ozdemiray

Altıngövde

2015

Asso for Info Science & Tech

View full text Add to dashboard Cite

Ranking methods are used in all aspects of life, from Google searches to sports tournaments. Because all ranking methods necessarily have advantages and disadvantages, USA Climbing, the organizer of national climbing competitions in the United States, changed their ranking method three times between 2009 and 2016. The combined rank method employed in 2015 marked a drastic step away from the previous two in that it failed to meet the independence of irrelevant alternatives (IIA) criterion and was almost impossible for spectators to use to calculate ranks on their own. We compare this more recent rank aggregation method with older USA Climbing score aggregation methods as well as other methods from the literature. Three particularly important methods we consider are (i) the combined rank method, (ii) a combination of the previous two USA Climbing score aggregation methods (the merged method), and (iii) a linear programming (LP)-based rank aggregation method from the literature. Using data from the 2016 Bouldering Youth National Championships, we perform leave-one-out cross validation and the Friedman hypothesis test to conclude that at the 99% confidence level, the LP-based rank aggregation method has significantly more predictive power than the other two methods, while there was insufficient evidence to distinguish between the predictive power of the combined rank method and the merged method. However, due to the desirable properties, such as the IIA criterion, satisfied by the merged method, we recommend this method for use in competitive climbing.

show abstract

Federated Search

Cited by 121 publications

References 204 publications

Federated Entity Search Using On-the-Fly Consolidation

Federated Entity Search Using On-the-Fly Consolidation

To what problem is distributed information retrieval the solution?

Explicit search result diversification using score and rank aggregation methods

Contact Info

Product

Resources

About