Distance between Sets

Levandowsky, M.; Winter, David J.

doi:10.1038/physci235060b0

Cited by 133 publications

(158 citation statements)

References 2 publications

Supporting

Mentioning

158

Contrasting

Order By: Relevance

“…JI', a variation of Jaccard Index [13], complements LD in order to avoid inaccurate decisions regarding the strings containing same words in different order.…”

Section: Utilization Of Ji'mentioning

confidence: 99%

Utilization of synergetic human-machine clouds: a big data cleaning case

İren

Kul

Bilgen

2014

Proceedings of the 1st International Workshop on CrowdSourcing in Software Engineering

View full text Add to dashboard Cite

Cloud computing and crowdsourcing are growing trends in IT. Combining the strengths of both machine and human clouds within a hybrid design enables us to overcome certain problems and achieve efficiencies. In this paper we present a case in which we developed a hybrid, throw-away prototype software system to solve a big data cleaning problem in which we corrected and normalized a data set of 53,822 academic publication records. The first step in our solution consists of utilization of external DOI query web services to label the records with matching DOIs. Then we used customized string similarity calculation algorithms based on Levensthein Distance and Jaccard Index to grade the similarity between records. Finally we used crowdsourcing to identify duplicates among the residual record set consisting of similar yet not identical records. We consider this proof of concept to be successful and report that we achieved certain results that we could not have achieved by using either human or machine clouds alone.

show abstract

“…JI', a variation of Jaccard Index [13], complements LD in order to avoid inaccurate decisions regarding the strings containing same words in different order.…”

Section: Utilization Of Ji'mentioning

confidence: 99%

Utilization of synergetic human-machine clouds: a big data cleaning case

İren

Kul

Bilgen

2014

Proceedings of the 1st International Workshop on CrowdSourcing in Software Engineering

View full text Add to dashboard Cite

show abstract

“…Given two non-empty sets, it measures the probability that an element of at least one of two sets is an element of both, and thus is a reasonable measure of similarity or overlap between the two [18]. The Jaccard Coefficient is defined as follows:…”

Section: Jaccardmentioning

confidence: 99%

Unsupervised Distance Learning by Rank Correlation Measures for Image Retrieval

Okada

Pedronette

Torres

2015

Proceedings of the 5th ACM on International Conference on Multimedia Retrieval

View full text Add to dashboard Cite

Ranking accurately collection images is the main objective of Content-based Image Retrieval (CBIR) systems. In fact, the set of images ranked at the first positions generally defines the effectiveness of provided search services, i.e., they are used for assessing automatically the quality of search systems as this set usually contains the collection images that are of interest. Recently, the use of ranking information (e.g., rank correlation) has been used in different research initiatives with the objective of improving the effectiveness of image retrieval tasks. This paper presents a broad rank correlation analysis for unsupervised distance learning on image retrieval tasks. Various well-known rank correlation measures are considered and two new measures are proposed. Several experiments were conducted considering various image datasets involving shape, color, and texture descriptors. Experimental results demonstrate that ranking information can be exploited for distance learning tasks successfully. Evaluated approaches yield better results in terms of effectiveness than various state-of-the-art algorithms.

show abstract

“…When time series are converted into sets, their similarity can be computed with the Jaccard metric [19], which is one of the most popular similarities for sets. It is computed as follows: Jaccard(S, Q) = |S∩Q| |S∪Q| .…”

Section: The Sts3 Algorithmmentioning

confidence: 99%

Set-based Similarity Search for Time Series

Peng

Wang

et al. 2016

Proceedings of the 2016 International Conference on Management of Data

View full text Add to dashboard Cite

A fundamental problem with time series involves k nearest neighbor (k-NN) query processing. However, existing methods are not fast enough for large datasets. In this paper, we propose a novel approach, STS3, for processing k-NN queries that transforms time series into sets and measures the similarity under the Jaccard metric. Our approach is more accurate than Dynamic Time Warping (DTW) in some suitable scenarios and is faster than most existing methods, due to the efficiency of similarity search for sets. In addition, we developed indexing, pruning and approximation techniques to improve the k-NN querying procedure. As shown in the experimental results, these techniques could accelerate the query processing effectively.

show abstract

Distance between Sets

Cited by 133 publications

References 2 publications

Utilization of synergetic human-machine clouds: a big data cleaning case

Utilization of synergetic human-machine clouds: a big data cleaning case

Unsupervised Distance Learning by Rank Correlation Measures for Image Retrieval

Set-based Similarity Search for Time Series

Contact Info

Product

Resources

About