Danish Asylum Adjudication using Deep Neural Networks and Natural Language Processing

Satya Mahesh Muddamsetty; Mohammad Naser Sabet Jahromi; Thomas B. Moeslund; Thomas Gammeltoft-Hansen

Danish Asylum Adjudication using Deep Neural Networks and Natural Language Processing

Satya Mahesh Muddamsetty^*, Mohammad Naser Sabet Jahromi, Thomas B. Moeslund, Thomas Gammeltoft-Hansen

^*Corresponding author for this work

MOBILE - Center for Global Mobility Law

Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review

Abstract

The Danish asylum adjudication procedure is a two-tiered system, with the Immigration Service making initial determinations and the Danish Refugee Appeals Board (RAB) automatically appealing cases that are rejected. This study aims to employ a deep neural network(DNN)based Natural Language Processing (NLP) pipeline to predict asylum decision-making outcomes using a dataset of over 15,515 Danish asylum decisions provided by the Danish Refugee Appeals Board (RAB) between January 1995 and January 2021. This research seeks to improve the performance and effectiveness of decision-making in asylum cases by addressing key challenges, such as modeling the asylum decision-making problem using NLP-based DNNs and dealing with class imbalance issues. Our preliminary results indicate that DNN-based NLP predictive models are capable of learning meaningful representations of asylum cases with high precision and recall, particularly when class weights are considered than the baseline DNN model.

Original language	English
Title of host publication	Proceedings of the Seventeenth International Workshop on Juris-Informatics 2023
Number of pages	14
Publication date	2023
Pages	92-105
Publication status	Published - 2023
Event	International Workshop on Juris-Informatics - Kumamoto, Japan Duration: 5 Jun 2023 → 6 Jun 2023 https://research.nii.ac.jp/~ksatoh/jurisin2023/

Conference

Conference	International Workshop on Juris-Informatics
Country/Territory	Japan
City	Kumamoto
Period	05/06/2023 → 06/06/2023
Internet address	https://research.nii.ac.jp/~ksatoh/jurisin2023/

Access to Document

Library access?

Check availability at the Royal Danish Library

Cite this

Muddamsetty, S. M., Jahromi, M. N. S., Moeslund, T. B., & Gammeltoft-Hansen, T. (2023). Danish Asylum Adjudication using Deep Neural Networks and Natural Language Processing. In Proceedings of the Seventeenth International Workshop on Juris-Informatics 2023 (pp. 92-105) https://vbn.aau.dk/en/publications/danish-asylum-adjudication-using-deep-neural-networks-and-natural

Danish Asylum Adjudication using Deep Neural Networks and Natural Language Processing. / Muddamsetty, Satya Mahesh; Jahromi, Mohammad Naser Sabet; Moeslund, Thomas B. et al.
Proceedings of the Seventeenth International Workshop on Juris-Informatics 2023. 2023. p. 92-105.

Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review

Muddamsetty, SM, Jahromi, MNS, Moeslund, TB & Gammeltoft-Hansen, T 2023, Danish Asylum Adjudication using Deep Neural Networks and Natural Language Processing. in Proceedings of the Seventeenth International Workshop on Juris-Informatics 2023. pp. 92-105, International Workshop on Juris-Informatics, Kumamoto, Japan, 05/06/2023. <https://vbn.aau.dk/en/publications/danish-asylum-adjudication-using-deep-neural-networks-and-natural>

@inproceedings{a17ee606d37645509302dff88eb31e3f,

title = "Danish Asylum Adjudication using Deep Neural Networks and Natural Language Processing",

abstract = "The Danish asylum adjudication procedure is a two-tiered system, with the Immigration Service making initial determinations and the Danish Refugee Appeals Board (RAB) automatically appealing cases that are rejected. This study aims to employ a deep neural network(DNN)based Natural Language Processing (NLP) pipeline to predict asylum decision-making outcomes using a dataset of over 15,515 Danish asylum decisions provided by the Danish Refugee Appeals Board (RAB) between January 1995 and January 2021. This research seeks to improve the performance and effectiveness of decision-making in asylum cases by addressing key challenges, such as modeling the asylum decision-making problem using NLP-based DNNs and dealing with class imbalance issues. Our preliminary results indicate that DNN-based NLP predictive models are capable of learning meaningful representations of asylum cases with high precision and recall, particularly when class weights are considered than the baseline DNN model.",

author = "Muddamsetty, {Satya Mahesh} and Jahromi, {Mohammad Naser Sabet} and Moeslund, {Thomas B.} and Thomas Gammeltoft-Hansen",

year = "2023",

language = "English",

pages = "92--105",

booktitle = "Proceedings of the Seventeenth International Workshop on Juris-Informatics 2023",

note = "International Workshop on Juris-Informatics ; Conference date: 05-06-2023 Through 06-06-2023",

url = "https://research.nii.ac.jp/~ksatoh/jurisin2023/",

}

TY - GEN

T1 - Danish Asylum Adjudication using Deep Neural Networks and Natural Language Processing

AU - Muddamsetty, Satya Mahesh

AU - Jahromi, Mohammad Naser Sabet

AU - Moeslund, Thomas B.

AU - Gammeltoft-Hansen, Thomas

PY - 2023

Y1 - 2023

N2 - The Danish asylum adjudication procedure is a two-tiered system, with the Immigration Service making initial determinations and the Danish Refugee Appeals Board (RAB) automatically appealing cases that are rejected. This study aims to employ a deep neural network(DNN)based Natural Language Processing (NLP) pipeline to predict asylum decision-making outcomes using a dataset of over 15,515 Danish asylum decisions provided by the Danish Refugee Appeals Board (RAB) between January 1995 and January 2021. This research seeks to improve the performance and effectiveness of decision-making in asylum cases by addressing key challenges, such as modeling the asylum decision-making problem using NLP-based DNNs and dealing with class imbalance issues. Our preliminary results indicate that DNN-based NLP predictive models are capable of learning meaningful representations of asylum cases with high precision and recall, particularly when class weights are considered than the baseline DNN model.

AB - The Danish asylum adjudication procedure is a two-tiered system, with the Immigration Service making initial determinations and the Danish Refugee Appeals Board (RAB) automatically appealing cases that are rejected. This study aims to employ a deep neural network(DNN)based Natural Language Processing (NLP) pipeline to predict asylum decision-making outcomes using a dataset of over 15,515 Danish asylum decisions provided by the Danish Refugee Appeals Board (RAB) between January 1995 and January 2021. This research seeks to improve the performance and effectiveness of decision-making in asylum cases by addressing key challenges, such as modeling the asylum decision-making problem using NLP-based DNNs and dealing with class imbalance issues. Our preliminary results indicate that DNN-based NLP predictive models are capable of learning meaningful representations of asylum cases with high precision and recall, particularly when class weights are considered than the baseline DNN model.

M3 - Article in proceedings

SP - 92

EP - 105

BT - Proceedings of the Seventeenth International Workshop on Juris-Informatics 2023

T2 - International Workshop on Juris-Informatics

Y2 - 5 June 2023 through 6 June 2023

ER -