Link to the University of Pittsburgh Homepage
Link to the University Library System Homepage Link to the Contact Us Form

LOCALIZATION OF EVENTS USING UNDERDEVELOPED MICROBLOGGING DATA

Anjum, Usman (2022) LOCALIZATION OF EVENTS USING UNDERDEVELOPED MICROBLOGGING DATA. Doctoral Dissertation, University of Pittsburgh. (Unpublished)

[img]
Preview
PDF (FINAL VERSION)
Updated Version

Download (5MB) | Preview

Abstract

Event localization is the task of finding the location of an event. Events are defined as significant one-time occurrences that show notable deviation from expected or normal behavior. Event localization has been studied in many domains including medical data, internet-of-things (IoT), sensor data, and microblogging/social media domain.
In this dissertation, we focus on event localization in the microblogging domain. The data in the microblogging presents a unique challenge in that it is underdeveloped. Underdeveloped data has low reliability and sporadic delivery slate. Since, microblogging data is underdeveloped it provides subjective and incomplete information, which is unsuitable for event localization. We propose enrichment methods for underdeveloped data that would make the data more suitable for event localization. Our enrichment methods include disaggregation, semantic filtering, and data generation using top-down and bottom-up approaches. Once the data is enriched, we identify event signatures that are specific to an event. We find both explicit and latent event signatures within the enriched data. Using these signatures an event can be efficiently localized. We use generated data and data collected from Twitter to test our enrichment methods and implement event localization strategies.


Share

Citation/Export:
Social Networking:
Share |

Details

Item Type: University of Pittsburgh ETD
Status: Unpublished
Creators/Authors:
CreatorsEmailPitt UsernameORCID
Anjum, Usmanusa3@pitt.eduusa30000-0002-9280-772X
ETD Committee:
TitleMemberEmail AddressPitt UsernameORCID
Thesis AdvisorKrishnamurthy, Prashantprashk@pitt.eduPRASHK
Committee CoChairZadorozhny, Vladimirviz@pitt.eduviz0000-0001-6420-1926
Committee MemberWeiss, Martin BHmbw@pitt.eduMBW0000-0001-6785-0913
Committee MemberAbdelhakim, Maimaia@pitt.eduMAIA0000-0001-8442-0974
Date: 17 January 2022
Date Type: Publication
Defense Date: 6 August 2021
Approval Date: 17 January 2022
Submission Date: 11 October 2021
Access Restriction: No restriction; Release the ETD for access worldwide immediately.
Number of Pages: 117
Institution: University of Pittsburgh
Schools and Programs: School of Information Sciences > Telecommunications
Degree: PhD - Doctor of Philosophy
Thesis Type: Doctoral Dissertation
Refereed: Yes
Uncontrolled Keywords: Event Localization, Event Detection, Datafusion, Disaggregation, AI, Agent-Based Modelling, Data Augmentation
Date Deposited: 17 Jan 2022 14:59
Last Modified: 17 Jan 2022 14:59
URI: http://d-scholarship.pitt.edu/id/eprint/41846

Metrics

Monthly Views for the past 3 years

Plum Analytics


Actions (login required)

View Item View Item