{"id":"https://openalex.org/W7134171015","doi":"https://doi.org/10.1109/bigdata66926.2025.11400789","title":"Leveraging Synthetically Generated Data for Real Estate Document Classification","display_name":"Leveraging Synthetically Generated Data for Real Estate Document Classification","publication_year":2025,"publication_date":"2025-12-08","ids":{"openalex":"https://openalex.org/W7134171015","doi":"https://doi.org/10.1109/bigdata66926.2025.11400789"},"language":"en","primary_location":{"id":"doi:10.1109/bigdata66926.2025.11400789","is_oa":false,"landing_page_url":"https://doi.org/10.1109/bigdata66926.2025.11400789","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE International Conference on Big Data (BigData)","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5003368121","display_name":"Tobias Deu\u00dfer","orcid":"https://orcid.org/0000-0003-4685-0847"},"institutions":[{"id":"https://openalex.org/I135140700","display_name":"University of Bonn","ror":"https://ror.org/041nas322","country_code":"DE","type":"education","lineage":["https://openalex.org/I135140700"]}],"countries":["DE"],"is_corresponding":true,"raw_author_name":"Tobias Deu\u00dfer","raw_affiliation_strings":["University of Bonn,Bonn,Germany"],"affiliations":[{"raw_affiliation_string":"University of Bonn,Bonn,Germany","institution_ids":["https://openalex.org/I135140700"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5128359152","display_name":"Gregor Ramien","orcid":null},"institutions":[{"id":"https://openalex.org/I133845707","display_name":"Energie Baden-W\u00fcrttemberg (Germany)","ror":"https://ror.org/041shdj17","country_code":"DE","type":"company","lineage":["https://openalex.org/I133845707"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Gregor Ramien","raw_affiliation_strings":["Atruvia AG,Karlsruhe,Germany"],"affiliations":[{"raw_affiliation_string":"Atruvia AG,Karlsruhe,Germany","institution_ids":["https://openalex.org/I133845707"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5128352473","display_name":"Nico Weber","orcid":null},"institutions":[{"id":"https://openalex.org/I133845707","display_name":"Energie Baden-W\u00fcrttemberg (Germany)","ror":"https://ror.org/041shdj17","country_code":"DE","type":"company","lineage":["https://openalex.org/I133845707"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Nico Weber","raw_affiliation_strings":["Atruvia AG,Karlsruhe,Germany"],"affiliations":[{"raw_affiliation_string":"Atruvia AG,Karlsruhe,Germany","institution_ids":["https://openalex.org/I133845707"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5040862341","display_name":"Maximilian Meidinger","orcid":null},"institutions":[{"id":"https://openalex.org/I133845707","display_name":"Energie Baden-W\u00fcrttemberg (Germany)","ror":"https://ror.org/041shdj17","country_code":"DE","type":"company","lineage":["https://openalex.org/I133845707"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Maximilian Meidinger","raw_affiliation_strings":["Atruvia AG,Karlsruhe,Germany"],"affiliations":[{"raw_affiliation_string":"Atruvia AG,Karlsruhe,Germany","institution_ids":["https://openalex.org/I133845707"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5119252625","display_name":"Max Hahnb\u00fcck","orcid":null},"institutions":[{"id":"https://openalex.org/I4210144576","display_name":"Fraunhofer Institute for Intelligent Analysis and Information Systems","ror":"https://ror.org/04nc32781","country_code":"DE","type":"facility","lineage":["https://openalex.org/I4210144576","https://openalex.org/I4923324"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Max Hahnb\u00fcck","raw_affiliation_strings":["Fraunhofer IAIS,Sankt Augustin,Germany"],"affiliations":[{"raw_affiliation_string":"Fraunhofer IAIS,Sankt Augustin,Germany","institution_ids":["https://openalex.org/I4210144576"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5128365886","display_name":"Christian Bauckhage","orcid":null},"institutions":[{"id":"https://openalex.org/I135140700","display_name":"University of Bonn","ror":"https://ror.org/041nas322","country_code":"DE","type":"education","lineage":["https://openalex.org/I135140700"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Christian Bauckhage","raw_affiliation_strings":["University of Bonn,Bonn,Germany"],"affiliations":[{"raw_affiliation_string":"University of Bonn,Bonn,Germany","institution_ids":["https://openalex.org/I135140700"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5128342889","display_name":"Rafet Sifa","orcid":null},"institutions":[{"id":"https://openalex.org/I135140700","display_name":"University of Bonn","ror":"https://ror.org/041nas322","country_code":"DE","type":"education","lineage":["https://openalex.org/I135140700"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Rafet Sifa","raw_affiliation_strings":["University of Bonn,Bonn,Germany"],"affiliations":[{"raw_affiliation_string":"University of Bonn,Bonn,Germany","institution_ids":["https://openalex.org/I135140700"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":7,"corresponding_author_ids":["https://openalex.org/A5003368121"],"corresponding_institution_ids":["https://openalex.org/I135140700"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.73159931,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"7253","last_page":"7257"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10601","display_name":"Handwritten Text Recognition Techniques","score":0.15549999475479126,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T10601","display_name":"Handwritten Text Recognition Techniques","score":0.15549999475479126,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T11550","display_name":"Text and Document Classification Technologies","score":0.08969999849796295,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T11006","display_name":"BIM and Construction Integration","score":0.06840000301599503,"subfield":{"id":"https://openalex.org/subfields/2215","display_name":"Building and Construction"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/real-estate","display_name":"Real estate","score":0.4034999907016754},{"id":"https://openalex.org/keywords/identification","display_name":"Identification (biology)","score":0.2897000014781952},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.28209999203681946},{"id":"https://openalex.org/keywords/feature","display_name":"Feature (linguistics)","score":0.26030001044273376},{"id":"https://openalex.org/keywords/information-system","display_name":"Information system","score":0.2549999952316284}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6064000129699707},{"id":"https://openalex.org/C82279013","wikidata":"https://www.wikidata.org/wiki/Q684740","display_name":"Real estate","level":2,"score":0.4034999907016754},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.36239999532699585},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.34790000319480896},{"id":"https://openalex.org/C116834253","wikidata":"https://www.wikidata.org/wiki/Q2039217","display_name":"Identification (biology)","level":2,"score":0.2897000014781952},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.28209999203681946},{"id":"https://openalex.org/C23123220","wikidata":"https://www.wikidata.org/wiki/Q816826","display_name":"Information retrieval","level":1,"score":0.27390000224113464},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.26030001044273376},{"id":"https://openalex.org/C180198813","wikidata":"https://www.wikidata.org/wiki/Q121182","display_name":"Information system","level":2,"score":0.2549999952316284},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.24269999563694}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1109/bigdata66926.2025.11400789","is_oa":false,"landing_page_url":"https://doi.org/10.1109/bigdata66926.2025.11400789","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE International Conference on Big Data (BigData)","raw_type":"proceedings-article"},{"id":"pmh:doi:10.48565/bonndoc-809","is_oa":false,"landing_page_url":"https://hdl.handle.net/20.500.11811/13972","pdf_url":null,"source":{"id":"https://openalex.org/S4406922384","display_name":"Open MIND","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Conferenceproceeding"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320311649","display_name":"Ministry of Education","ror":"https://ror.org/036nq5137"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":27,"referenced_works":["https://openalex.org/W1497131193","https://openalex.org/W2757311323","https://openalex.org/W2973794125","https://openalex.org/W3092415316","https://openalex.org/W3093601344","https://openalex.org/W3103291112","https://openalex.org/W3115467802","https://openalex.org/W3160861144","https://openalex.org/W4249906708","https://openalex.org/W4287855183","https://openalex.org/W4313007903","https://openalex.org/W4317905944","https://openalex.org/W4360764576","https://openalex.org/W4388994251","https://openalex.org/W4388994351","https://openalex.org/W4389888517","https://openalex.org/W4390754928","https://openalex.org/W4391093139","https://openalex.org/W4391094171","https://openalex.org/W4400654973","https://openalex.org/W4401283900","https://openalex.org/W4402684073","https://openalex.org/W4406458719","https://openalex.org/W4406460142","https://openalex.org/W4408859653","https://openalex.org/W4411630019","https://openalex.org/W4416677830"],"related_works":[],"abstract_inverted_index":{"Document":[0],"classification":[1],"in":[2,103,125],"regulated":[3],"domains":[4],"like":[5],"law,":[6],"finance,":[7],"or":[8],"real":[9,57,101],"estate":[10,58],"is":[11],"hindered":[12],"by":[13],"the":[14,98,108,120],"scarcity":[15],"of":[16,39,100,110],"labeled":[17],"data":[18,32,46,113],"and":[19,45,63,71,87,118],"strict":[20],"privacy":[21],"constraints.":[22],"This":[23],"paper":[24],"presents":[25],"a":[26,37,80],"pipeline":[27],"for":[28,33,122],"synthetically":[29],"generating":[30],"training":[31],"document":[34,53],"classifiers":[35],"using":[36,75,111],"combination":[38],"domain-specific":[40],"templates,":[41],"large":[42],"language":[43],"models,":[44],"augmentation":[47],"techniques.":[48],"Focusing":[49],"on":[50,83,90],"two":[51],"key":[52],"types":[54],"relevant":[55],"to":[56,114],"workflows,":[59],"<em>Child":[60],"Support":[61],"Certificate":[62],"Refurbishment":[64],"Roadmap</em>,":[65],"we":[66],"construct":[67],"realistic":[68],"multi-page":[69],"documents":[70,102],"generate":[72],"negative":[73],"classes":[74],"LLM-generated":[76],"distractors.":[77],"We":[78],"train":[79],"BERT-based":[81],"classifier":[82],"this":[84],"synthetic":[85,112],"dataset":[86],"evaluate":[88],"it":[89],"real-world":[91],"OCR-extracted":[92],"documents,":[93],"achieving":[94],"strong":[95],"performance":[96],"despite":[97],"absence":[99],"training.":[104],"Our":[105],"findings":[106],"highlight":[107],"feasibility":[109],"overcome":[115],"annotation":[116],"bottlenecks":[117],"pave":[119],"way":[121],"broader":[123],"applications":[124],"privacy-sensitive":[126],"industries.":[127]},"counts_by_year":[],"updated_date":"2026-04-09T08:11:56.329763","created_date":"2026-03-09T00:00:00"}