[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-127554-en":3,"doc-seo-127554-105":30,"detail-sidebar-cat-0-en-105":92},{"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":20,"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},127554,687207017582,"Himbo","https://ap-avatar.wpscdn.com/davatar_994ba38a5ba835b3df7d355c54d3ed8d",8,"Research & Report","Using the Interest Theory of Rights and Hohfeldian Taxonomy to Address a Gap in Machine Learning Methods for Legal Document Analysis","Legal documents embody rights and duties, and their analysis using machine learning is often limited to surface tasks such as named entity recognition, sentiment analysis, or court case classification. The study argues that without modeling these core legal features, automated assessment becomes a bottleneck. It proposes using legal theory—specifically the interest theory of rights and the first-order Hohfeldian taxonomy—to derive structured dimensions suitable for ML abstractions. A philosophy-informed heuristic with language models identifies key Hohfeldian relations with high classification accuracy on UK religious discrimination policy texts.","ARTICLE   \n [https://doi.org/10.1057/s41599-023-01693-z](https://doi.org/10.1057/s41599-023-01693-z)  OPEN  \nUsing the interest theory of rights and Hohfeldian taxonomy to address a gap in machine learning methods for legal document analysis  \nAhmed Izzidien  1 ✉  \nRights and duties are essential features of legal documents. Machine learning algorithms have been increasingly applied to extract information from such texts. Currently, their main focus is on named entity recognition, sentiment analysis, and the classification of court cases to predict court outcome. In this paper it is argued that until the essential features of such texts are captured, their analysis can remain bottle-necked by the very technology being used to assess them. As such, the use of legal theory to identify the most pertinent dimensions of such texts is proposed. Specifically, the interest theory of rights, and the first-order Hohfeldian taxonomy of legal relations. These principal legal dimensions allow for a stratified representation of knowledge, making them ideal for the abstractions needed for machine learning. This study considers how such dimensions may be identified. To do so it implements a novel heuristic based in philosophy coupled with language models. Hohfeldian relations of ‘ rights-duties’ vs. ‘privileges-no-rights’ are determined to be identifiable. Classification of each type of relation to accuracies of 92.5% is found using Sentence Bidirectional Encoder Representations from Transformers. Testing is carried out on religious discrimination policy texts in the United Kingdom.  \n1 The Faculty of Law, The University of Cambridge, Cambridge, UK. ✉ email: [ai297@cam.ac.uk](ai297@cam.ac.uk)  \nHUMANITIES AND SOCIAL SCIENCES COMMUNICATIONS | (2023)10:251 | [https://doi.org/10.1057/s41599-023-01693-z](https://doi.org/10.1057/s41599-023-01693-z) 1  \nIntroduction  \nA  \nrtificial intelligence (AI), and specifically Natural Language Processing (NLP) through Machine Learning (ML), is increasingly being utilised in public interest technolo-  \ngies, such as in government departments, courts, and NGOs offering legal services (de Sousa et al., 2019) . Using a systematic protocol for literature reviews and meta-analysis (PRISMA) a study (de Sousa et al., 2019) demonstrated that managers of public organisations have considerably increased the adoption of AI-based systems to improve efficiency (Mehr et al., 2017) . For example, it has been found that a month’s work at the US Department for Labour can be completed in a day with higher accuracy. Machines take into account factors at orders of magnitude greater than people without tiring. Indeed, this human frailty, when not checked, has been found to be a factor that negatively impacts decisions, such as court rulings (Danziger et al., 2011) .  \nIt has further been found that implementing ML in such contexts has made procedures more efficient. The Supreme Court of Brazil found AI contributed positively to its procedural speed, for example (de Sousa et al., 2022) . Big data NLP also allows for a systematic analysis of documents at scale, allowing for new findings to materialise which are typically not possible with the human eye (Nay, 2018) . Indeed, almost all law is expressed in natural language; therefore NLP may be said to be a key component to understand and predict law at scale (Nay, 2018) .  \nLegal documents may be characterised as essentially texts which describe power interactions in society with consideration to the outcomes of such interactions (Boswell and Smith, 2017; Oliver and Cairney, 2019) . These power interactions typically exist to further the interests of one or more parties involved (Hewitt, 2009; Michael et al., 2011) .  \nBased on the nature of these interests, rights and duties and their concomitant legal relations become imposed (Kramer, 2001, 2017) . Given that policy documents, which find themselves into legislation through a well-documented process (Parliament, 2021) are based on a considerati","cbCaim3Gt8mUlWsP","https://ap.wps.com/l/cbCaim3Gt8mUlWsP","pdf",1368604,1,15,"English","en",105,"# Introduction\n## Legal documents as structured power and interest relations\n## Machine learning limitations and need for feature capture\n## Proposed legal-theory-driven dimensions for ML","[{\"question\":\"What gap in machine learning for legal document analysis does the paper address?\",\"answer\":\"It targets the gap where current ML applications focus on tasks like NER, sentiment, and case classification without explicitly capturing essential legal features such as rights and duties.\"},{\"question\":\"Which legal frameworks are used to define the features for machine learning?\",\"answer\":\"The paper uses the interest theory of rights and the first-order Hohfeldian taxonomy of legal relations to build a structured representation of legal knowledge.\"},{\"question\":\"How are Hohfeldian relations identified and evaluated?\",\"answer\":\"A philosophy-based heuristic combined with language models is used to determine types of Hohfeldian relations, and classification performance is tested with reported accuracies up to 92.5% on UK policy texts.\"}]","Using the Interest Theory of Rights and Hohfeldian Taxonomy to Address a Gap in Machine Learning Methods for Legal Document Analysis | PDF",1785939929,38,{"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":87,"head_meta":89,"extra_data":91,"updated_unix":28},"using-the-interest-theory-of-rights-and-hohfeldian-taxonomy-to-address-a-gap-in-machine-learning-methods-for-legal-document-analysis","",{"@graph":36,"@context":86},[37,54,69],{"@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/using-the-interest-theory-of-rights-and-hohfeldian-taxonomy-to-address-a-gap-in-machine-learning-methods-for-legal-document-analysis/127554/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":63,"encodingFormat":61,"isAccessibleForFree":64,"interactionStatistic":65},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-22","2026-08-05",true,{"@type":66,"interactionType":67,"userInteractionCount":20},"InteractionCounter",{"@type":68},"ViewAction",{"@type":70,"mainEntity":71},"FAQPage",[72,78,82],{"name":73,"@type":74,"acceptedAnswer":75},"What gap in machine learning for legal document analysis does the paper address?","Question",{"text":76,"@type":77},"It targets the gap where current ML applications focus on tasks like NER, sentiment, and case classification without explicitly capturing essential legal features such as rights and duties.","Answer",{"name":79,"@type":74,"acceptedAnswer":80},"Which legal frameworks are used to define the features for machine learning?",{"text":81,"@type":77},"The paper uses the interest theory of rights and the first-order Hohfeldian taxonomy of legal relations to build a structured representation of legal knowledge.",{"name":83,"@type":74,"acceptedAnswer":84},"How are Hohfeldian relations identified and evaluated?",{"text":85,"@type":77},"A philosophy-based heuristic combined with language models is used to determine types of Hohfeldian relations, and classification performance is tested with reported accuracies up to 92.5% on UK policy texts.","https://schema.org",{"og:url":52,"og:type":88,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":90,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":93},[94,98,102,106,111,116,121,124,129,132,136],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":95,"show_sort_weight":96,"slug":97},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":99,"show_sort_weight":100,"slug":101},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":103,"show_sort_weight":104,"slug":105},"Exam",70,"exam",{"id":107,"doc_module":4,"doc_module_name":46,"category_name":108,"show_sort_weight":109,"slug":110},5,"Comic",60,"comic",{"id":112,"doc_module":4,"doc_module_name":46,"category_name":113,"show_sort_weight":114,"slug":115},6,"Technology",50,"technology",{"id":117,"doc_module":4,"doc_module_name":46,"category_name":118,"show_sort_weight":119,"slug":120},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":122,"slug":123},30,"research-report",{"id":125,"doc_module":4,"doc_module_name":46,"category_name":126,"show_sort_weight":127,"slug":128},9,"Religion & Spirituality",20,"religion-spirituality",{"id":127,"doc_module":4,"doc_module_name":46,"category_name":130,"show_sort_weight":127,"slug":131},"World Cup","world-cup",{"id":133,"doc_module":4,"doc_module_name":46,"category_name":134,"show_sort_weight":133,"slug":135},10,"Lifestyle","lifestyle",{"id":137,"doc_module":4,"doc_module_name":46,"category_name":138,"show_sort_weight":107,"slug":139},19,"General","general"]