Crystal Cube: Forecasting Disruptive Events

Anna L. Buczak; Benjamin D. Baugher; Christine S. Martin; Meg W. Keiley-Listermann; James Howard; Nathan H. Parrish; Anton Q. Stalick; Daniel S. Berman; Mark H. Dredze

doi:10.1080/08839514.2021.2001179

Crystal Cube: Forecasting Disruptive Events

Anna L. Buczak, Benjamin D. Baugher, Christine S. Martin, Meg W. Keiley-Listermann, James Howard, Nathan H. Parrish, Anton Q. Stalick, Daniel S. Berman, Mark H. Dredze

Research output: Contribution to journal › Article › peer-review

Abstract

Disruptive events within a country can have global repercussions, creating a need for the anticipation and planning of these events. Crystal Cube (CC) is a novel approach to forecasting disruptive political events at least one month into the future. The system uses a recurrent neural network and a novel measure of event similarity between past and current events. We also introduce the innovative Thermometer of Irregular Leadership Change (ILC). We present an evaluation of CC in predicting ILC for 167 countries and show promising results in forecasting events one to twelve months in advance. We compare CC results with results using a random forest as well as previous work.

Original language	English (US)
Article number	2001179
Journal	Applied Artificial Intelligence
Volume	36
Issue number	1
DOIs	https://doi.org/10.1080/08839514.2021.2001179
State	Published - 2022
Externally published	Yes

ASJC Scopus subject areas

Artificial Intelligence

Access to Document

10.1080/08839514.2021.2001179

Cite this

@article{8c430a663a5f4edc8b054819a2981d74,

title = "Crystal Cube: Forecasting Disruptive Events",

abstract = "Disruptive events within a country can have global repercussions, creating a need for the anticipation and planning of these events. Crystal Cube (CC) is a novel approach to forecasting disruptive political events at least one month into the future. The system uses a recurrent neural network and a novel measure of event similarity between past and current events. We also introduce the innovative Thermometer of Irregular Leadership Change (ILC). We present an evaluation of CC in predicting ILC for 167 countries and show promising results in forecasting events one to twelve months in advance. We compare CC results with results using a random forest as well as previous work.",

author = "Buczak, {Anna L.} and Baugher, {Benjamin D.} and Martin, {Christine S.} and Keiley-Listermann, {Meg W.} and James Howard and Parrish, {Nathan H.} and Stalick, {Anton Q.} and Berman, {Daniel S.} and Dredze, {Mark H.}",

note = "Publisher Copyright: {\textcopyright} 2022 JHU APL. Published with license by Taylor & Francis Group, LLC.",

year = "2022",

doi = "10.1080/08839514.2021.2001179",

language = "English (US)",

volume = "36",

journal = "Applied Artificial Intelligence",

issn = "0883-9514",

publisher = "Taylor and Francis Ltd.",

number = "1",

}

TY - JOUR

T1 - Crystal Cube

T2 - Forecasting Disruptive Events

AU - Buczak, Anna L.

AU - Baugher, Benjamin D.

AU - Martin, Christine S.

AU - Keiley-Listermann, Meg W.

AU - Howard, James

AU - Parrish, Nathan H.

AU - Stalick, Anton Q.

AU - Berman, Daniel S.

AU - Dredze, Mark H.

PY - 2022

Y1 - 2022

N2 - Disruptive events within a country can have global repercussions, creating a need for the anticipation and planning of these events. Crystal Cube (CC) is a novel approach to forecasting disruptive political events at least one month into the future. The system uses a recurrent neural network and a novel measure of event similarity between past and current events. We also introduce the innovative Thermometer of Irregular Leadership Change (ILC). We present an evaluation of CC in predicting ILC for 167 countries and show promising results in forecasting events one to twelve months in advance. We compare CC results with results using a random forest as well as previous work.

AB - Disruptive events within a country can have global repercussions, creating a need for the anticipation and planning of these events. Crystal Cube (CC) is a novel approach to forecasting disruptive political events at least one month into the future. The system uses a recurrent neural network and a novel measure of event similarity between past and current events. We also introduce the innovative Thermometer of Irregular Leadership Change (ILC). We present an evaluation of CC in predicting ILC for 167 countries and show promising results in forecasting events one to twelve months in advance. We compare CC results with results using a random forest as well as previous work.

UR - http://www.scopus.com/inward/record.url?scp=85119286190&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85119286190&partnerID=8YFLogxK

U2 - 10.1080/08839514.2021.2001179

DO - 10.1080/08839514.2021.2001179

M3 - Article

AN - SCOPUS:85119286190

SN - 0883-9514

VL - 36

JO - Applied Artificial Intelligence

JF - Applied Artificial Intelligence

IS - 1

M1 - 2001179

ER -

Crystal Cube: Forecasting Disruptive Events

Abstract

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this