Link to the University of Pittsburgh Homepage
Link to the University Library System Homepage Link to the Contact Us Form
D-Scholarship @ Pitt Banner and Links to Homepage

Causal Discovery in Social Weather System

He, Jiexiao (2019) Causal Discovery in Social Weather System. Master's Thesis, University of Pittsburgh. (Unpublished)

[img]
Preview
 PDF
Download (7MB) | Preview

Abstract

In this thesis, we explore relationships between social variables that can be used to assess social events and conditions (social weather). Social weather may have hidden causes, and social events may be related to several variables distributed over diverse data sources. We built an infrastructure integrating data sources using the World Bank, stock market, happiness, terrorism, and Internet usage in the US. We performed data cleaning, data normalization and data aggregation and implemented our data store using Influx DB; we used Grafana for visual exploration of the integrated database.
The integrated data store allowed us to explore data trends and relationships among the social variables. In particular, we explored causal links among large amount of social variables using the Tetrad framework. We applied several causal discovery algorithms to generate a support matrix and used the voting approach to find strong causal relationships. We demonstrated that our causal discovery results are consistent with a well-known economic model.

Share

Citation/Export:
Social Networking:
Share |

Details

Item Type: University of Pittsburgh ETD
Status: Unpublished
Creators/Authors:
CreatorsEmailPitt UsernameORCID
He, JiexiaoJIH102@pitt.edujih1020000-0002-2706-7788
ETD Committee:
TitleMemberEmail AddressPitt UsernameORCID
Committee ChairZadorozhny, Vladimirviz@pitt.edu
Committee MemberMunro, Paulpwm@pitt.edu
Committee MemberDruzdzel, Marekdruzdzel@pitt.edu
Date: 22 May 2019
Date Type: Publication
Defense Date: 18 March 2019
Approval Date: 22 May 2019
Submission Date: 8 April 2019
Access Restriction: No restriction; Release the ETD for access worldwide immediately.
Number of Pages: 65
Institution: University of Pittsburgh
Schools and Programs: School of Computing and Information > Information Science
Degree: MSIS - Master of Science in Information Science
Thesis Type: Master's Thesis
Refereed: Yes
Uncontrolled Keywords: Time-series System, Causal Discovery, Majority Vote.
Related URLs:
Date Deposited: 22 May 2019 12:35
Last Modified: 22 May 2019 12:35
URI: http://d-scholarship.pitt.edu/id/eprint/36409

Metrics

Monthly Views for the past 3 years

Plum Analytics

Actions (login required)

View Item View Item