Efficient estimation for high similarities using odd sketches

Michael Mitzenmacher, Rasmus Pagh, Ninh Dang Pham

Research output: Conference Article in Proceeding or Book/Report chapterArticle in proceedingsResearchpeer-review


Estimating set similarity is a central problem in many computer applications. In this paper we introduce the Odd Sketch, a compact binary sketch for estimating the Jaccard similarity of two sets. The exclusive-or of two sketches equals the sketch of the symmetric difference of the two sets. This means that Odd Sketches provide a highly space-efficient estimator for sets of high similarity, which is relevant in applications such as web duplicate detection, collaborative filtering, and association rule learning. The method extends to weighted Jaccard similarity, relevant e.g. for TF-IDF vector comparison. We present a theoretical analysis of the quality of estimation to guarantee the reliability of Odd Sketch-based estimators. Our experiments confirm this efficiency, and demonstrate the efficiency of Odd Sketches in comparison with $b$-bit minwise hashing schemes on association rule learning and web duplicate detection tasks.
Original languageEnglish
Title of host publicationProceedings of the 23rd international conference on World wide web : WWW '14
Number of pages10
PublisherAssociation for Computing Machinery
Publication date2014
ISBN (Electronic)978-1-4503-2744-2
Publication statusPublished - 2014


  • Set Similarity
  • Odd Sketch
  • Jaccard Similarity
  • Symmetric Difference
  • Web Duplicate Detection
  • Collaborative Filtering
  • Association Rule Learning
  • Weighted Jaccard Similarity
  • TF-IDF Vector Comparison
  • Minwise Hashing


Dive into the research topics of 'Efficient estimation for high similarities using odd sketches'. Together they form a unique fingerprint.
  • Best Paper Award

    Pham, N. D. (Participant)

    7 Apr 201411 Apr 2014

    Activity: Other activity typesOther (prizes, external teaching and other activities) - Prizes, scholarships, distinctions

Cite this