Temporal Ordinance Mining for Event-Driven Social Media Reaction Analytics

Aparna S. Varde, Gerard De Melo, Boxiang Dong

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

1 Scopus citations

Abstract

As a growing number of policies are adopted to address the substantial rise in urbanization, there is a significant push for smart governance, endowing transparency in decision-making and enabling greater public involvement. The thriving concept of smart governance goes beyond just cities, ultimately aiming at a smart planet. Ordinances (local laws) affect our life with regard to health, business, etc. This is particularly notable during major events such as the recent pandemic, which may lead to rapid changes in ordinances, pertaining for instance to public safety, disaster management, and recovery phases. However, many citizens view ordinances as impervious and complex. This position paper proposes a research agenda enabling novel forms of ordinance content analysis over time and temporal web question answering (QA) for both legislators and the broader public. Along with this, we aim to analyze social media posts so as to track the public opinion before and after the introduction of ordinances. Challenges include addressing concepts changing over time and infusing subtle human reasoning in mining, which we aim to address by harnessing terminology evolution methods and commonsense knowledge sources, respectively. We aim to make the results of the historical ordinance mining and event-driven analysis seamlessly accessible, relying on a robust semantic understanding framework to flexibly support web QA.

Original languageEnglish
Title of host publicationACM Web Conference 2023 - Companion of the World Wide Web Conference, WWW 2023
PublisherAssociation for Computing Machinery, Inc
Pages1225-1227
Number of pages3
ISBN (Electronic)9781450394161
DOIs
StatePublished - 30 Apr 2023
Event2023 World Wide Web Conference, WWW 2023 - Austin, United States
Duration: 30 Apr 20234 May 2023

Publication series

NameACM Web Conference 2023 - Companion of the World Wide Web Conference, WWW 2023

Conference

Conference2023 World Wide Web Conference, WWW 2023
Country/TerritoryUnited States
CityAustin
Period30/04/234/05/23

Keywords

  • Commonsense knowledge
  • historical data
  • local laws
  • machine learning
  • NLP
  • smart governance
  • social media
  • terminology evolution
  • text mining
  • urban policy
  • web Q&A

Fingerprint

Dive into the research topics of 'Temporal Ordinance Mining for Event-Driven Social Media Reaction Analytics'. Together they form a unique fingerprint.

Cite this