[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-127283-en":3,"doc-seo-127283-105":30,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":4,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":14,"update_tm":28,"read_time":29},127283,2336475104736,"Quinn","https://ap-avatar.wpscdn.com/avatar/22000c4c5e0e5b17e70?x-image-process=image/resize,m_fixed,w_180,h_180&k=1786591360781797222",8,"Research & Report","Explainability of machine learning approaches in forensic linguistics: a case study in geolinguistic authorship profiling","Forensic authorship profiling uses linguistic markers to infer characteristics of an author behind a text. The same idea parallels dialect classification, where a model predicts a text’s linguistic variety from its own language. Because forensic linguistics often lacks transparency, explainable machine learning is still uncommon in legal contexts. This study examines explainability for variety classification to support geolinguistic profiling of unknown German-speaking social media authorship, identifying the most impactful lexical items and showing reliance on place names.","Explainability of machine learning approaches in forensic linguistics: a case study in geolinguistic authorship profiling  \nDana Roemling  \nUniversity of Birmingham University of Helsinki  \n[d.roemling@bham.ac.uk](d.roemling@bham.ac.uk)  \nYves Scherrer  \nUniversity of Oslo  \nUniversity of Helsinki [yves.scherrer@ifi.uio.no](yves.scherrer@ifi.uio.no)  \nAleksandra Mileti  \nUniversity of Helsinki University Sorbonne Nouvelle  \n[aleksandra.miletic@helsinki.fi](aleksandra.miletic@helsinki.fi)  \narXiv :2404 . 185 10v2 [ cs .CL] 1 Jul 2024  \nAbstract  \nForensic authorship profiling uses linguistic markers to infer characteristics about an author of a text. This task is paralleled in dialect classification, where a prediction is made about the linguistic variety of a text based on the text itself. While there have been significant advances in recent years in variety classification, forensic linguistics rarely relies on these approaches due to their lack of transparency, among other reasons. In this paper we therefore explore the explainability of machine learning approaches considering the forensic context. We focus on variety classification as a means of geolinguistic profiling of unknown texts based on social media data from the German-speaking area. For this, we identify the lexical items that are the most impactful for the variety classification. We find that the extracted lexical features are indeed representative of their respective varieties and note that the trained models also rely on place names for classifications.  \n1 Introduction  \nForensic authorship analysis is a key area of research within forensic linguistics that seeks to gain information about the author(s) of a text. Generally, there are two central domains of research within authorship analysis: comparative authorship analysis uses various methodologies to compare questioned and known documents to attribute authorship, while sociolinguistic or authorship profiling relies on the analysis of questioned texts alone to infer characteristics of an author (Grant, 2022 ; Roemling and Grieve, 2024) . Both areas of authorship analysis can be approached qualitatively and/or quantitatively if the amount of available data permits it. For example, quantitative work in authorship profiling has researched inferring age or gender (Nini, 2018) or native language influence (Kredens et al., 2019) from questioned documents.  \nNevertheless, forensic authorship profiling is often carried out in a manual or qualitative way, relying on the expertise of the forensic linguist. This is evident in famous examples like the Unabomber analysis (Leonard et al., 2017) or the devil strip ransom note (Shuy, 2001) . In both cases, law enforcement was interested in the regional origin of the authors. This background can be inferred through analyzing the regional linguistic variation, i.e., the use of regional dialect, in the questioned documents. This is referred to as regional or geolinguistic profiling (Roemling and Grieve, 2024) and is a task parallel to inferring the regional variety of a document as is done in language identification (Jauhiainen et al., 2019) .  \nEven though research in forensic linguistics works more and more with statistical and computational approaches (e.g., Bevendorff et al., 2023 ; Ishihara et al., 2024 ; Nini et al., 2024), authorship profiling often remains a manual task. This is at times credited to the black-box approaches in current NLP research, meaning that the lack of explainability precludes these approaches from being used in legal settings (see Nini, 2023) .  \n2 Related work  \nThe interest in explainability of machine learning (and in particular, neural) approaches is not only a relevant research area for forensic linguistics. Previous work, including on language identification, has focused on understanding how classifiers come to their predictions. Research started by creating an interpretable model around the actual classification approach to explain the predi","cbCaism8MTLGzvMp","https://ap.wps.com/l/cbCaism8MTLGzvMp","pdf",303064,1,7,"English","en",105,"# Introduction\n# Related work","[{\"question\":\"What problem does forensic authorship profiling address?\",\"answer\":\"It infers characteristics about an author of a questioned text using linguistic markers. The study frames this alongside dialect classification, where linguistic variety is predicted from text.\"},{\"question\":\"Why is explainability important in forensic NLP applications?\",\"answer\":\"Forensic contexts often require transparency, and black-box approaches may be unsuitable because limited explainability can prevent adoption in legal settings.\"},{\"question\":\"How does the paper approach explainability for geolinguistic authorship profiling?\",\"answer\":\"It focuses on variety classification for German-speaking social media data and identifies the lexical items most impactful for the classification. It also observes that trained models rely on place names.\"}]","Explainability of machine learning approaches in forensic linguistics: a case study in geolinguistic authorship profiling | PDF",1785938094,18,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"explainability-of-machine-learning-approaches-in-forensic-linguistics-a-case-study-in-geolinguistic-authorship-profiling","",{"@graph":36,"@context":85},[37,54,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/explainability-of-machine-learning-approaches-in-forensic-linguistics-a-case-study-in-geolinguistic-authorship-profiling/127283/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":62,"encodingFormat":61,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-05",true,{"@type":65,"interactionType":66,"userInteractionCount":4},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What problem does forensic authorship profiling address?","Question",{"text":75,"@type":76},"It infers characteristics about an author of a questioned text using linguistic markers. The study frames this alongside dialect classification, where linguistic variety is predicted from text.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"Why is explainability important in forensic NLP applications?",{"text":80,"@type":76},"Forensic contexts often require transparency, and black-box approaches may be unsuitable because limited explainability can prevent adoption in legal settings.",{"name":82,"@type":73,"acceptedAnswer":83},"How does the paper approach explainability for geolinguistic authorship profiling?",{"text":84,"@type":76},"It focuses on variety classification for German-speaking social media data and identifies the lexical items most impactful for the classification. It also observes that trained models rely on place names.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,119,122,127,130,134],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":21,"doc_module":4,"doc_module_name":46,"category_name":116,"show_sort_weight":117,"slug":118},"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":120,"slug":121},30,"research-report",{"id":123,"doc_module":4,"doc_module_name":46,"category_name":124,"show_sort_weight":125,"slug":126},9,"Religion & Spirituality",20,"religion-spirituality",{"id":125,"doc_module":4,"doc_module_name":46,"category_name":128,"show_sort_weight":125,"slug":129},"World Cup","world-cup",{"id":131,"doc_module":4,"doc_module_name":46,"category_name":132,"show_sort_weight":131,"slug":133},10,"Lifestyle","lifestyle",{"id":135,"doc_module":4,"doc_module_name":46,"category_name":136,"show_sort_weight":106,"slug":137},19,"General","general"]