Extracting Social Connections from Finnish Karelian Refugee Interviews Using LLMs Article Swipe
Joonatan Laato
,
Jenna Kanerva
,
John Loehr
,
Virpi Lummaa
,
Filip Ginter
·
YOU?
·
· 2025
· Open Access
·
· DOI: https://doi.org/10.48550/arxiv.2502.13566
YOU?
·
· 2025
· Open Access
·
· DOI: https://doi.org/10.48550/arxiv.2502.13566
We performed a zero-shot information extraction study on a historical collection of 89,339 brief Finnish-language interviews of refugee families relocated post-WWII from Finnish Eastern Karelia. Our research objective is two-fold. First, we aim to extract social organizations and hobbies from the free text of the interviews, separately for each family member. These can act as a proxy variable indicating the degree of social integration of refugees in their new environment. Second, we aim to evaluate several alternative ways to approach this task, comparing a number of generative models and a supervised learning approach, to gain a broader insight into the relative merits of these different approaches and their applicability in similar studies. We find that the best generative model (GPT-4) is roughly on par with human performance, at an F-score of 88.8%. Interestingly, the best open generative model (Llama-3-70B-Instruct) reaches almost the same performance, at 87.7% F-score, demonstrating that open models are becoming a viable alternative for some practical tasks even on non-English data. Additionally, we test a supervised learning alternative, where we fine-tune a Finnish BERT model (FinBERT) using GPT-4 generated training data. By this method, we achieved an F-score of 84.1% already with 6K interviews up to an F-score of 86.3% with 30k interviews. Such an approach would be particularly appealing in cases where the computational resources are limited, or there is a substantial mass of data to process.
Related Topics
Concepts
Metadata
- Type
- preprint
- Language
- en
- Landing Page
- http://arxiv.org/abs/2502.13566
- https://arxiv.org/pdf/2502.13566
- OA Status
- green
- Related Works
- 10
- OpenAlex ID
- https://openalex.org/W4407764437
All OpenAlex metadata
Raw OpenAlex JSON
- OpenAlex ID
-
https://openalex.org/W4407764437Canonical identifier for this work in OpenAlex
- DOI
-
https://doi.org/10.48550/arxiv.2502.13566Digital Object Identifier
- Title
-
Extracting Social Connections from Finnish Karelian Refugee Interviews Using LLMsWork title
- Type
-
preprintOpenAlex work type
- Language
-
enPrimary language
- Publication year
-
2025Year of publication
- Publication date
-
2025-02-19Full publication date if available
- Authors
-
Joonatan Laato, Jenna Kanerva, John Loehr, Virpi Lummaa, Filip GinterList of authors in order
- Landing page
-
https://arxiv.org/abs/2502.13566Publisher landing page
- PDF URL
-
https://arxiv.org/pdf/2502.13566Direct link to full text PDF
- Open access
-
YesWhether a free full text is available
- OA status
-
greenOpen access status per OpenAlex
- OA URL
-
https://arxiv.org/pdf/2502.13566Direct OA link when available
- Concepts
-
Refugee, Political science, LawTop concepts (fields/topics) attached by OpenAlex
- Cited by
-
0Total citation count in OpenAlex
- Related works (count)
-
10Other works algorithmically related by OpenAlex
Full payload
| id | https://openalex.org/W4407764437 |
|---|---|
| doi | https://doi.org/10.48550/arxiv.2502.13566 |
| ids.doi | https://doi.org/10.48550/arxiv.2502.13566 |
| ids.openalex | https://openalex.org/W4407764437 |
| fwci | |
| type | preprint |
| title | Extracting Social Connections from Finnish Karelian Refugee Interviews Using LLMs |
| biblio.issue | |
| biblio.volume | |
| biblio.last_page | |
| biblio.first_page | |
| topics[0].id | https://openalex.org/T13910 |
| topics[0].field.id | https://openalex.org/fields/33 |
| topics[0].field.display_name | Social Sciences |
| topics[0].score | 0.8306000232696533 |
| topics[0].domain.id | https://openalex.org/domains/2 |
| topics[0].domain.display_name | Social Sciences |
| topics[0].subfield.id | https://openalex.org/subfields/3300 |
| topics[0].subfield.display_name | General Social Sciences |
| topics[0].display_name | Computational and Text Analysis Methods |
| topics[1].id | https://openalex.org/T11719 |
| topics[1].field.id | https://openalex.org/fields/18 |
| topics[1].field.display_name | Decision Sciences |
| topics[1].score | 0.8116999864578247 |
| topics[1].domain.id | https://openalex.org/domains/2 |
| topics[1].domain.display_name | Social Sciences |
| topics[1].subfield.id | https://openalex.org/subfields/1803 |
| topics[1].subfield.display_name | Management Science and Operations Research |
| topics[1].display_name | Data Quality and Management |
| is_xpac | False |
| apc_list | |
| apc_paid | |
| concepts[0].id | https://openalex.org/C173145845 |
| concepts[0].level | 2 |
| concepts[0].score | 0.8650132417678833 |
| concepts[0].wikidata | https://www.wikidata.org/wiki/Q131572 |
| concepts[0].display_name | Refugee |
| concepts[1].id | https://openalex.org/C17744445 |
| concepts[1].level | 0 |
| concepts[1].score | 0.38666167855262756 |
| concepts[1].wikidata | https://www.wikidata.org/wiki/Q36442 |
| concepts[1].display_name | Political science |
| concepts[2].id | https://openalex.org/C199539241 |
| concepts[2].level | 1 |
| concepts[2].score | 0.0 |
| concepts[2].wikidata | https://www.wikidata.org/wiki/Q7748 |
| concepts[2].display_name | Law |
| keywords[0].id | https://openalex.org/keywords/refugee |
| keywords[0].score | 0.8650132417678833 |
| keywords[0].display_name | Refugee |
| keywords[1].id | https://openalex.org/keywords/political-science |
| keywords[1].score | 0.38666167855262756 |
| keywords[1].display_name | Political science |
| language | en |
| locations[0].id | pmh:oai:arXiv.org:2502.13566 |
| locations[0].is_oa | True |
| locations[0].source.id | https://openalex.org/S4306400194 |
| locations[0].source.issn | |
| locations[0].source.type | repository |
| locations[0].source.is_oa | True |
| locations[0].source.issn_l | |
| locations[0].source.is_core | False |
| locations[0].source.is_in_doaj | False |
| locations[0].source.display_name | arXiv (Cornell University) |
| locations[0].source.host_organization | https://openalex.org/I205783295 |
| locations[0].source.host_organization_name | Cornell University |
| locations[0].source.host_organization_lineage | https://openalex.org/I205783295 |
| locations[0].license | |
| locations[0].pdf_url | https://arxiv.org/pdf/2502.13566 |
| locations[0].version | submittedVersion |
| locations[0].raw_type | text |
| locations[0].license_id | |
| locations[0].is_accepted | False |
| locations[0].is_published | False |
| locations[0].raw_source_name | |
| locations[0].landing_page_url | http://arxiv.org/abs/2502.13566 |
| locations[1].id | doi:10.48550/arxiv.2502.13566 |
| locations[1].is_oa | True |
| locations[1].source.id | https://openalex.org/S4306400194 |
| locations[1].source.issn | |
| locations[1].source.type | repository |
| locations[1].source.is_oa | True |
| locations[1].source.issn_l | |
| locations[1].source.is_core | False |
| locations[1].source.is_in_doaj | False |
| locations[1].source.display_name | arXiv (Cornell University) |
| locations[1].source.host_organization | https://openalex.org/I205783295 |
| locations[1].source.host_organization_name | Cornell University |
| locations[1].source.host_organization_lineage | https://openalex.org/I205783295 |
| locations[1].license | cc-by |
| locations[1].pdf_url | |
| locations[1].version | |
| locations[1].raw_type | article |
| locations[1].license_id | https://openalex.org/licenses/cc-by |
| locations[1].is_accepted | False |
| locations[1].is_published | |
| locations[1].raw_source_name | |
| locations[1].landing_page_url | https://doi.org/10.48550/arxiv.2502.13566 |
| indexed_in | arxiv, datacite |
| authorships[0].author.id | https://openalex.org/A5116339286 |
| authorships[0].author.orcid | |
| authorships[0].author.display_name | Joonatan Laato |
| authorships[0].author_position | first |
| authorships[0].raw_author_name | Laato, Joonatan |
| authorships[0].is_corresponding | False |
| authorships[1].author.id | https://openalex.org/A5048932608 |
| authorships[1].author.orcid | https://orcid.org/0000-0003-4580-5366 |
| authorships[1].author.display_name | Jenna Kanerva |
| authorships[1].author_position | middle |
| authorships[1].raw_author_name | Kanerva, Jenna |
| authorships[1].is_corresponding | False |
| authorships[2].author.id | https://openalex.org/A5089727131 |
| authorships[2].author.orcid | |
| authorships[2].author.display_name | John Loehr |
| authorships[2].author_position | middle |
| authorships[2].raw_author_name | Loehr, John |
| authorships[2].is_corresponding | False |
| authorships[3].author.id | https://openalex.org/A5073396537 |
| authorships[3].author.orcid | https://orcid.org/0000-0002-2128-7587 |
| authorships[3].author.display_name | Virpi Lummaa |
| authorships[3].author_position | middle |
| authorships[3].raw_author_name | Lummaa, Virpi |
| authorships[3].is_corresponding | False |
| authorships[4].author.id | https://openalex.org/A5019929457 |
| authorships[4].author.orcid | https://orcid.org/0000-0002-5484-6103 |
| authorships[4].author.display_name | Filip Ginter |
| authorships[4].author_position | last |
| authorships[4].raw_author_name | Ginter, Filip |
| authorships[4].is_corresponding | False |
| has_content.pdf | False |
| has_content.grobid_xml | False |
| is_paratext | False |
| open_access.is_oa | True |
| open_access.oa_url | https://arxiv.org/pdf/2502.13566 |
| open_access.oa_status | green |
| open_access.any_repository_has_fulltext | False |
| created_date | 2025-10-10T00:00:00 |
| display_name | Extracting Social Connections from Finnish Karelian Refugee Interviews Using LLMs |
| has_fulltext | False |
| is_retracted | False |
| updated_date | 2025-11-06T06:51:31.235846 |
| primary_topic.id | https://openalex.org/T13910 |
| primary_topic.field.id | https://openalex.org/fields/33 |
| primary_topic.field.display_name | Social Sciences |
| primary_topic.score | 0.8306000232696533 |
| primary_topic.domain.id | https://openalex.org/domains/2 |
| primary_topic.domain.display_name | Social Sciences |
| primary_topic.subfield.id | https://openalex.org/subfields/3300 |
| primary_topic.subfield.display_name | General Social Sciences |
| primary_topic.display_name | Computational and Text Analysis Methods |
| related_works | https://openalex.org/W4391375266, https://openalex.org/W2899084033, https://openalex.org/W2748952813, https://openalex.org/W2034193224, https://openalex.org/W2477007413, https://openalex.org/W2753976929, https://openalex.org/W2006402215, https://openalex.org/W2760152342, https://openalex.org/W1541164599, https://openalex.org/W2282708004 |
| cited_by_count | 0 |
| locations_count | 2 |
| best_oa_location.id | pmh:oai:arXiv.org:2502.13566 |
| best_oa_location.is_oa | True |
| best_oa_location.source.id | https://openalex.org/S4306400194 |
| best_oa_location.source.issn | |
| best_oa_location.source.type | repository |
| best_oa_location.source.is_oa | True |
| best_oa_location.source.issn_l | |
| best_oa_location.source.is_core | False |
| best_oa_location.source.is_in_doaj | False |
| best_oa_location.source.display_name | arXiv (Cornell University) |
| best_oa_location.source.host_organization | https://openalex.org/I205783295 |
| best_oa_location.source.host_organization_name | Cornell University |
| best_oa_location.source.host_organization_lineage | https://openalex.org/I205783295 |
| best_oa_location.license | |
| best_oa_location.pdf_url | https://arxiv.org/pdf/2502.13566 |
| best_oa_location.version | submittedVersion |
| best_oa_location.raw_type | text |
| best_oa_location.license_id | |
| best_oa_location.is_accepted | False |
| best_oa_location.is_published | False |
| best_oa_location.raw_source_name | |
| best_oa_location.landing_page_url | http://arxiv.org/abs/2502.13566 |
| primary_location.id | pmh:oai:arXiv.org:2502.13566 |
| primary_location.is_oa | True |
| primary_location.source.id | https://openalex.org/S4306400194 |
| primary_location.source.issn | |
| primary_location.source.type | repository |
| primary_location.source.is_oa | True |
| primary_location.source.issn_l | |
| primary_location.source.is_core | False |
| primary_location.source.is_in_doaj | False |
| primary_location.source.display_name | arXiv (Cornell University) |
| primary_location.source.host_organization | https://openalex.org/I205783295 |
| primary_location.source.host_organization_name | Cornell University |
| primary_location.source.host_organization_lineage | https://openalex.org/I205783295 |
| primary_location.license | |
| primary_location.pdf_url | https://arxiv.org/pdf/2502.13566 |
| primary_location.version | submittedVersion |
| primary_location.raw_type | text |
| primary_location.license_id | |
| primary_location.is_accepted | False |
| primary_location.is_published | False |
| primary_location.raw_source_name | |
| primary_location.landing_page_url | http://arxiv.org/abs/2502.13566 |
| publication_date | 2025-02-19 |
| publication_year | 2025 |
| referenced_works_count | 0 |
| abstract_inverted_index.a | 2, 8, 55, 83, 89, 95, 153, 167, 174, 224 |
| abstract_inverted_index.6K | 195 |
| abstract_inverted_index.By | 184 |
| abstract_inverted_index.We | 0, 112 |
| abstract_inverted_index.an | 128, 189, 199, 207 |
| abstract_inverted_index.as | 54 |
| abstract_inverted_index.at | 127, 144 |
| abstract_inverted_index.be | 210 |
| abstract_inverted_index.in | 66, 109, 213 |
| abstract_inverted_index.is | 28, 120, 223 |
| abstract_inverted_index.of | 11, 16, 43, 61, 64, 85, 102, 130, 191, 201, 227 |
| abstract_inverted_index.on | 7, 122, 161 |
| abstract_inverted_index.or | 221 |
| abstract_inverted_index.to | 33, 73, 78, 93, 198, 229 |
| abstract_inverted_index.up | 197 |
| abstract_inverted_index.we | 31, 71, 165, 172, 187 |
| abstract_inverted_index.30k | 204 |
| abstract_inverted_index.Our | 25 |
| abstract_inverted_index.act | 53 |
| abstract_inverted_index.aim | 32, 72 |
| abstract_inverted_index.and | 37, 88, 106 |
| abstract_inverted_index.are | 151, 219 |
| abstract_inverted_index.can | 52 |
| abstract_inverted_index.for | 47, 156 |
| abstract_inverted_index.new | 68 |
| abstract_inverted_index.par | 123 |
| abstract_inverted_index.the | 40, 44, 59, 99, 115, 133, 141, 216 |
| abstract_inverted_index.BERT | 176 |
| abstract_inverted_index.Such | 206 |
| abstract_inverted_index.best | 116, 134 |
| abstract_inverted_index.data | 228 |
| abstract_inverted_index.each | 48 |
| abstract_inverted_index.even | 160 |
| abstract_inverted_index.find | 113 |
| abstract_inverted_index.free | 41 |
| abstract_inverted_index.from | 21, 39 |
| abstract_inverted_index.gain | 94 |
| abstract_inverted_index.into | 98 |
| abstract_inverted_index.mass | 226 |
| abstract_inverted_index.open | 135, 149 |
| abstract_inverted_index.same | 142 |
| abstract_inverted_index.some | 157 |
| abstract_inverted_index.test | 166 |
| abstract_inverted_index.text | 42 |
| abstract_inverted_index.that | 114, 148 |
| abstract_inverted_index.this | 80, 185 |
| abstract_inverted_index.ways | 77 |
| abstract_inverted_index.with | 124, 194, 203 |
| abstract_inverted_index.84.1% | 192 |
| abstract_inverted_index.86.3% | 202 |
| abstract_inverted_index.87.7% | 145 |
| abstract_inverted_index.GPT-4 | 180 |
| abstract_inverted_index.These | 51 |
| abstract_inverted_index.brief | 13 |
| abstract_inverted_index.cases | 214 |
| abstract_inverted_index.data. | 163, 183 |
| abstract_inverted_index.human | 125 |
| abstract_inverted_index.model | 118, 137, 177 |
| abstract_inverted_index.proxy | 56 |
| abstract_inverted_index.study | 6 |
| abstract_inverted_index.task, | 81 |
| abstract_inverted_index.tasks | 159 |
| abstract_inverted_index.their | 67, 107 |
| abstract_inverted_index.there | 222 |
| abstract_inverted_index.these | 103 |
| abstract_inverted_index.using | 179 |
| abstract_inverted_index.where | 171, 215 |
| abstract_inverted_index.would | 209 |
| abstract_inverted_index.88.8%. | 131 |
| abstract_inverted_index.89,339 | 12 |
| abstract_inverted_index.First, | 30 |
| abstract_inverted_index.almost | 140 |
| abstract_inverted_index.degree | 60 |
| abstract_inverted_index.family | 49 |
| abstract_inverted_index.merits | 101 |
| abstract_inverted_index.models | 87, 150 |
| abstract_inverted_index.number | 84 |
| abstract_inverted_index.social | 35, 62 |
| abstract_inverted_index.viable | 154 |
| abstract_inverted_index.(GPT-4) | 119 |
| abstract_inverted_index.Eastern | 23 |
| abstract_inverted_index.F-score | 129, 190, 200 |
| abstract_inverted_index.Finnish | 22, 175 |
| abstract_inverted_index.Second, | 70 |
| abstract_inverted_index.already | 193 |
| abstract_inverted_index.broader | 96 |
| abstract_inverted_index.extract | 34 |
| abstract_inverted_index.hobbies | 38 |
| abstract_inverted_index.insight | 97 |
| abstract_inverted_index.member. | 50 |
| abstract_inverted_index.method, | 186 |
| abstract_inverted_index.reaches | 139 |
| abstract_inverted_index.refugee | 17 |
| abstract_inverted_index.roughly | 121 |
| abstract_inverted_index.several | 75 |
| abstract_inverted_index.similar | 110 |
| abstract_inverted_index.F-score, | 146 |
| abstract_inverted_index.Karelia. | 24 |
| abstract_inverted_index.achieved | 188 |
| abstract_inverted_index.approach | 79, 208 |
| abstract_inverted_index.becoming | 152 |
| abstract_inverted_index.evaluate | 74 |
| abstract_inverted_index.families | 18 |
| abstract_inverted_index.learning | 91, 169 |
| abstract_inverted_index.limited, | 220 |
| abstract_inverted_index.process. | 230 |
| abstract_inverted_index.refugees | 65 |
| abstract_inverted_index.relative | 100 |
| abstract_inverted_index.research | 26 |
| abstract_inverted_index.studies. | 111 |
| abstract_inverted_index.training | 182 |
| abstract_inverted_index.variable | 57 |
| abstract_inverted_index.(FinBERT) | 178 |
| abstract_inverted_index.appealing | 212 |
| abstract_inverted_index.approach, | 92 |
| abstract_inverted_index.comparing | 82 |
| abstract_inverted_index.different | 104 |
| abstract_inverted_index.fine-tune | 173 |
| abstract_inverted_index.generated | 181 |
| abstract_inverted_index.objective | 27 |
| abstract_inverted_index.performed | 1 |
| abstract_inverted_index.post-WWII | 20 |
| abstract_inverted_index.practical | 158 |
| abstract_inverted_index.relocated | 19 |
| abstract_inverted_index.resources | 218 |
| abstract_inverted_index.two-fold. | 29 |
| abstract_inverted_index.zero-shot | 3 |
| abstract_inverted_index.approaches | 105 |
| abstract_inverted_index.collection | 10 |
| abstract_inverted_index.extraction | 5 |
| abstract_inverted_index.generative | 86, 117, 136 |
| abstract_inverted_index.historical | 9 |
| abstract_inverted_index.indicating | 58 |
| abstract_inverted_index.interviews | 15, 196 |
| abstract_inverted_index.separately | 46 |
| abstract_inverted_index.supervised | 90, 168 |
| abstract_inverted_index.alternative | 76, 155 |
| abstract_inverted_index.information | 4 |
| abstract_inverted_index.integration | 63 |
| abstract_inverted_index.interviews, | 45 |
| abstract_inverted_index.interviews. | 205 |
| abstract_inverted_index.non-English | 162 |
| abstract_inverted_index.substantial | 225 |
| abstract_inverted_index.alternative, | 170 |
| abstract_inverted_index.environment. | 69 |
| abstract_inverted_index.particularly | 211 |
| abstract_inverted_index.performance, | 126, 143 |
| abstract_inverted_index.Additionally, | 164 |
| abstract_inverted_index.applicability | 108 |
| abstract_inverted_index.computational | 217 |
| abstract_inverted_index.demonstrating | 147 |
| abstract_inverted_index.organizations | 36 |
| abstract_inverted_index.Interestingly, | 132 |
| abstract_inverted_index.Finnish-language | 14 |
| abstract_inverted_index.(Llama-3-70B-Instruct) | 138 |
| cited_by_percentile_year | |
| countries_distinct_count | 0 |
| institutions_distinct_count | 5 |
| citation_normalized_percentile |