TY - GEN
T1 - TURBaN: A Theory-Guided Model for Unemployment Rate Prediction Using Bayesian Network in Pandemic Scenario
AU - Das, Monidipa
AU - Basheer, Aysha
AU - Bandyopadhyay, Sanghamitra
N1 - Publisher Copyright:
© 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.
PY - 2023
Y1 - 2023
N2 - Unemployment rate is one of the key contributors that reflect the economic condition of a country. Accurate prediction of unemployment rate is a critically significant as well as demanding task which helps the government and the policymakers to make vital decisions. Though the recent research thrust is primarily towards hybridization of various linear and non-linear models, these may not perform satisfactorily well under the circumstances of unexpected events, e.g., during sudden outbreak of any infectious disease. In this paper, we explore this fact with respect to the current scenario of coronavirus disease (COVID) pandemic. Further, we show that explicit Bayesian modeling of pandemic impact on unemployment rate, together with theoretical insights from epidemiological models, can address this issue to some extent. Our developed theory-guided model for unemployment rate prediction using Bayesian network (TURBaN) is evaluated in terms of predicting unemployment rate in various states of India under COVID-19 pandemic scenario. The experimental result demonstrates the efficacy of TURBaN, which outperforms the state-of-the-art hybrid techniques in majority of the cases.
AB - Unemployment rate is one of the key contributors that reflect the economic condition of a country. Accurate prediction of unemployment rate is a critically significant as well as demanding task which helps the government and the policymakers to make vital decisions. Though the recent research thrust is primarily towards hybridization of various linear and non-linear models, these may not perform satisfactorily well under the circumstances of unexpected events, e.g., during sudden outbreak of any infectious disease. In this paper, we explore this fact with respect to the current scenario of coronavirus disease (COVID) pandemic. Further, we show that explicit Bayesian modeling of pandemic impact on unemployment rate, together with theoretical insights from epidemiological models, can address this issue to some extent. Our developed theory-guided model for unemployment rate prediction using Bayesian network (TURBaN) is evaluated in terms of predicting unemployment rate in various states of India under COVID-19 pandemic scenario. The experimental result demonstrates the efficacy of TURBaN, which outperforms the state-of-the-art hybrid techniques in majority of the cases.
KW - Bayesian network
KW - Epidemiology
KW - Theory-guided modeling
KW - Time series prediction
KW - Unemployment rate
UR - http://www.scopus.com/inward/record.url?scp=85163298428&partnerID=8YFLogxK
U2 - 10.1007/978-3-031-27409-1_47
DO - 10.1007/978-3-031-27409-1_47
M3 - Conference contribution
SN - 9783031274084
T3 - Lecture Notes in Networks and Systems
SP - 521
EP - 531
BT - Hybrid Intelligent Systems - 22nd International Conference on Hybrid Intelligent Systems HIS 2022
A2 - Abraham, Ajith
A2 - Abraham, Ajith
A2 - Hong, Tzung-Pei
A2 - Kotecha, Ketan
A2 - Ma, Kun
A2 - Manghirmalani Mishra, Pooja
A2 - Gandhi, Niketa
ER -
