Causality and uncertainty of information for content understanding

Adrienne Raglin, Raha Moraffah, Huan Liu

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Tasks require a clear picture of the context or the backdrop that frames the circumstances. Additionally tasks require a clear understanding of the content, the information available that will be used for completion of the task. Often the task involves a single or a set of decisions along the way. However, obtaining that content is not a perfect one. Understanding the content with is possible constraints, limitations, uncertainties adds to the challenge. To attempt to generate and express this the idea of an uncertainty of information concept that includes key aspects of causal reasoning is presented in this paper. In the paper the uncertainty of information (UoI) idea is discussed and how causality can be infused into this concept to not just provide another value for uncertainty be the causes. Moreover, can a causal UoI concept expand the idea so that a computational expression can capture the nuances of causal reasoning? This paper presents a possible vision.

Original languageEnglish (US)
Title of host publicationProceedings - 2020 IEEE 2nd International Conference on Cognitive Machine Intelligence, CogMI 2020
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages109-113
Number of pages5
ISBN (Electronic)9781728141442
DOIs
StatePublished - Oct 2020
Event2nd IEEE International Conference on Cognitive Machine Intelligence, CogMI 2020 - Virtual, Atlanta, United States
Duration: Dec 1 2020Dec 3 2020

Publication series

NameProceedings - 2020 IEEE 2nd International Conference on Cognitive Machine Intelligence, CogMI 2020

Conference

Conference2nd IEEE International Conference on Cognitive Machine Intelligence, CogMI 2020
Country/TerritoryUnited States
CityVirtual, Atlanta
Period12/1/2012/3/20

Keywords

  • Causality
  • Data preprocessing
  • Data transformation
  • Feature selection
  • Reasoning
  • Uncertainty

ASJC Scopus subject areas

  • Artificial Intelligence
  • Computer Science Applications
  • Software
  • Cognitive Neuroscience

Fingerprint

Dive into the research topics of 'Causality and uncertainty of information for content understanding'. Together they form a unique fingerprint.

Cite this