Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

Using Text-Based Life Trajectories from Swedish Register Data to Predict Residential Mobility with Pretrained Transformers

Stark, Philipp LU ; Sopasakis, Alexandros LU orcid ; Hall, Ola LU and Grillitsch, Markus LU orcid (2026) 30th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2026 In Lecture Notes in Computer Science 16600 LNAI. p.487-498
Abstract

We transform large-scale Swedish register data into textual life trajectories to address two long-standing challenges in data analysis: high cardinality of categorical variables and inconsistencies in coding schemes over time. Leveraging this uniquely comprehensive population register, we convert register data from 6.9 million individuals (2001–2013) into semantically rich texts and predict individuals’ residential mobility in later years (2013–2017). These life trajectories combine demographic information with annual changes in residence, work, education, income, and family circumstances, allowing us to assess how effectively such sequences support longitudinal prediction. We compare multiple NLP architectures (including LSTM,... (More)

We transform large-scale Swedish register data into textual life trajectories to address two long-standing challenges in data analysis: high cardinality of categorical variables and inconsistencies in coding schemes over time. Leveraging this uniquely comprehensive population register, we convert register data from 6.9 million individuals (2001–2013) into semantically rich texts and predict individuals’ residential mobility in later years (2013–2017). These life trajectories combine demographic information with annual changes in residence, work, education, income, and family circumstances, allowing us to assess how effectively such sequences support longitudinal prediction. We compare multiple NLP architectures (including LSTM, DistilBERT, BERT, and Qwen) and find that sequential and transformer-based models capture temporal and semantic structure more effectively than baseline models. The results show that textualized register data preserves meaningful information about individual pathways and supports complex, scalable modeling. Because few countries maintain longitudinal microdata with comparable coverage and precision, this dataset enables analyses and methodological tests that would be difficult or impossible elsewhere, offering a rigorous testbed for developing and evaluating new sequence-modeling approaches. Overall, our findings demonstrate that combining semantically rich register data with modern language models can substantially advance longitudinal analysis in social sciences.

(Less)
Please use this url to cite or link to this publication:
author
; ; and
organization
publishing date
type
Chapter in Book/Report/Conference proceeding
publication status
published
subject
keywords
Life Trajectory, Pretrained Transformer, Register Data, Residential Mobility, Sequence Modeling
host publication
Advances in Knowledge Discovery and Data Mining. PAKDD 2026
series title
Lecture Notes in Computer Science
editor
Wong, Raymond Chi-Wing ; Kwok, James ; Tong, Hanghang ; Lu, Hua ; Salim, Flora ; Song, Yuanfeng and Yiu, Man Lung
volume
16600 LNAI
pages
12 pages
publisher
Springer Science and Business Media B.V.
conference name
30th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2026
conference location
Hong Kong, China
conference dates
2026-06-09 - 2026-06-12
external identifiers
  • scopus:105041781262
ISSN
0302-9743
1611-3349
ISBN
9789819214679
DOI
10.1007/978-981-92-1468-6_34
language
English
LU publication?
yes
additional info
Publisher Copyright: © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2026.
id
dea896e6-99f0-4bd3-81cd-a079579df386
date added to LUP
2026-06-24 13:23:39
date last changed
2026-08-06 22:33:19
@inproceedings{dea896e6-99f0-4bd3-81cd-a079579df386,
  abstract     = {{<p>We transform large-scale Swedish register data into textual life trajectories to address two long-standing challenges in data analysis: high cardinality of categorical variables and inconsistencies in coding schemes over time. Leveraging this uniquely comprehensive population register, we convert register data from 6.9 million individuals (2001–2013) into semantically rich texts and predict individuals’ residential mobility in later years (2013–2017). These life trajectories combine demographic information with annual changes in residence, work, education, income, and family circumstances, allowing us to assess how effectively such sequences support longitudinal prediction. We compare multiple NLP architectures (including LSTM, DistilBERT, BERT, and Qwen) and find that sequential and transformer-based models capture temporal and semantic structure more effectively than baseline models. The results show that textualized register data preserves meaningful information about individual pathways and supports complex, scalable modeling. Because few countries maintain longitudinal microdata with comparable coverage and precision, this dataset enables analyses and methodological tests that would be difficult or impossible elsewhere, offering a rigorous testbed for developing and evaluating new sequence-modeling approaches. Overall, our findings demonstrate that combining semantically rich register data with modern language models can substantially advance longitudinal analysis in social sciences.</p>}},
  author       = {{Stark, Philipp and Sopasakis, Alexandros and Hall, Ola and Grillitsch, Markus}},
  booktitle    = {{Advances in Knowledge Discovery and Data Mining. PAKDD 2026}},
  editor       = {{Wong, Raymond Chi-Wing and Kwok, James and Tong, Hanghang and Lu, Hua and Salim, Flora and Song, Yuanfeng and Yiu, Man Lung}},
  isbn         = {{9789819214679}},
  issn         = {{0302-9743}},
  keywords     = {{Life Trajectory; Pretrained Transformer; Register Data; Residential Mobility; Sequence Modeling}},
  language     = {{eng}},
  month        = {{06}},
  pages        = {{487--498}},
  publisher    = {{Springer Science and Business Media B.V.}},
  series       = {{Lecture Notes in Computer Science}},
  title        = {{Using Text-Based Life Trajectories from Swedish Register Data to Predict Residential Mobility with Pretrained Transformers}},
  url          = {{http://dx.doi.org/10.1007/978-981-92-1468-6_34}},
  doi          = {{10.1007/978-981-92-1468-6_34}},
  volume       = {{16600 LNAI}},
  year         = {{2026}},
}