Surprisingness – A Novel Objective Interestingness Measure in Hypergraph Pattern Mining for Unsupervised Common Sense Learning

Shujing Ke, Pieter Spronck, Ben Goertzel, Alex van der Peet

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Abstract

Pattern mining usually results in huge amounts of patterns, among which only small percentages are interesting. In this paper, Surprisingness (including Surpringness_I and Surpringness_II) is proposed as an innovative objective multivariate interestingness measure for automatically identifying interesting patterns from a large quantity of patterns. Surprisingness is applicable in unstructured or semi-structured, multi-domain or mixed-domain data compared to existing measures. An experiment has been conducted enabling unsupervised learning of common sense, interesting patterns and exceptions from Wikipedia extracted data of random mixed domains (represented as a directed labeled hypergraphs), using the combinations of support and Surpringness.
Original languageEnglish
Title of host publicationProceedings of the ICBK 2021
Publication statusPublished - 2021
Event12th IEEE International Conference on Big Knowledge (ICBK-2021) - Auckland, New Zealand
Duration: 7 Dec 20218 Dec 2021

Conference

Conference12th IEEE International Conference on Big Knowledge (ICBK-2021)
Country/TerritoryNew Zealand
CityAuckland
Period7/12/218/12/21

Fingerprint

Dive into the research topics of 'Surprisingness – A Novel Objective Interestingness Measure in Hypergraph Pattern Mining for Unsupervised Common Sense Learning'. Together they form a unique fingerprint.

Cite this