Detecting Group Anomalies in Tera-Scale Multi-Aspect Data via Dense-Subtensor Mining

Cited 0 time in webofscience Cited 0 time in scopus
  • Hit : 147
  • Download : 12
DC FieldValueLanguage
dc.contributor.authorShin, Kijungko
dc.contributor.authorHooi, Bryanko
dc.contributor.authorKim, Jisuko
dc.contributor.authorFaloutsos, Christosko
dc.date.accessioned2021-08-03T02:10:20Z-
dc.date.available2021-08-03T02:10:20Z-
dc.date.created2021-08-03-
dc.date.created2021-08-03-
dc.date.created2021-08-03-
dc.date.created2021-08-03-
dc.date.created2021-08-03-
dc.date.created2021-08-03-
dc.date.created2021-08-03-
dc.date.issued2021-04-
dc.identifier.citationFRONTIERS IN BIG DATA, v.3, no.N, pp.58 - 58-
dc.identifier.issn2624-909X-
dc.identifier.urihttp://hdl.handle.net/10203/286976-
dc.description.abstractHow can we detect fraudulent lockstep behavior in large-scale multi-aspect data (i.e., tensors)? Can we detect it when data are too large to fit in memory or even on a disk? Past studies have shown that dense subtensors in real-world tensors (e.g., social media, Wikipedia, TCP dumps, etc.) signal anomalous or fraudulent behavior such as retweet boosting, bot activities, and network attacks. Thus, various approaches, including tensor decomposition and search, have been proposed for detecting dense subtensors rapidly and accurately. However, existing methods suffer from low accuracy, or they assume that tensors are small enough to fit in main memory, which is unrealistic in many real-world applications such as social media and web. To overcome these limitations, we propose D-Cube, a disk-based dense-subtensor detection method, which also can run in a distributed manner across multiple machines. Compared to state-of-the-art methods, D-Cube is (1) Memory Efficient: requires up to 1,561x less memory and handles 1,000x larger data (2.6TB), (2) Fast: up to 7x faster due to its near-linear scalability, (3) Provably Accurate: gives a guarantee on the densities of the detected subtensors, and (4) Effective: spotted network attacks from TCP dumps and synchronized behavior in rating data most accurately.-
dc.languageEnglish-
dc.publisherFRONTIERS MEDIA SA-
dc.titleDetecting Group Anomalies in Tera-Scale Multi-Aspect Data via Dense-Subtensor Mining-
dc.typeArticle-
dc.identifier.scopusid2-s2.0-85118717332-
dc.type.rimsART-
dc.citation.volume3-
dc.citation.issueN-
dc.citation.beginningpage58-
dc.citation.endingpage58-
dc.citation.publicationnameFRONTIERS IN BIG DATA-
dc.identifier.doi10.3389/fdata.2020.594302-
dc.contributor.localauthorShin, Kijung-
dc.contributor.nonIdAuthorHooi, Bryan-
dc.contributor.nonIdAuthorKim, Jisu-
dc.contributor.nonIdAuthorFaloutsos, Christos-
dc.description.isOpenAccessY-
dc.type.journalArticleArticle-
dc.subject.keywordAuthortensor-
dc.subject.keywordAuthordense subtensor-
dc.subject.keywordAuthoranomaly detection-
dc.subject.keywordAuthorfraud detection-
dc.subject.keywordAuthorout-of-core algorithm-
dc.subject.keywordAuthordistributed algorithm-
dc.subject.keywordPlusSUBGRAPH-
Appears in Collection
AI-Journal Papers(저널논문)
Files in This Item
121055.pdf(2.56 MB)Download

qr_code

  • mendeley

    citeulike


rss_1.0 rss_2.0 atom_1.0