[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-0-en-105":3,"doc-seo-148864-105":59,"doc-detail-148864-en":134},{"code":4,"msg":5,"data":6},0,"success",[7,13,18,23,28,33,38,43,48,51,55],{"id":8,"doc_module":4,"doc_module_name":9,"category_name":10,"show_sort_weight":11,"slug":12},1,"Document","Story & Novel",90,"story-novel",{"id":14,"doc_module":4,"doc_module_name":9,"category_name":15,"show_sort_weight":16,"slug":17},2,"Literature",80,"literature",{"id":19,"doc_module":4,"doc_module_name":9,"category_name":20,"show_sort_weight":21,"slug":22},4,"Exam",70,"exam",{"id":24,"doc_module":4,"doc_module_name":9,"category_name":25,"show_sort_weight":26,"slug":27},5,"Comic",60,"comic",{"id":29,"doc_module":4,"doc_module_name":9,"category_name":30,"show_sort_weight":31,"slug":32},6,"Technology",50,"technology",{"id":34,"doc_module":4,"doc_module_name":9,"category_name":35,"show_sort_weight":36,"slug":37},7,"Healthcare",40,"healthcare",{"id":39,"doc_module":4,"doc_module_name":9,"category_name":40,"show_sort_weight":41,"slug":42},8,"Research & Report",30,"research-report",{"id":44,"doc_module":4,"doc_module_name":9,"category_name":45,"show_sort_weight":46,"slug":47},9,"Religion & Spirituality",20,"religion-spirituality",{"id":46,"doc_module":4,"doc_module_name":9,"category_name":49,"show_sort_weight":46,"slug":50},"World Cup","world-cup",{"id":52,"doc_module":4,"doc_module_name":9,"category_name":53,"show_sort_weight":52,"slug":54},10,"Lifestyle","lifestyle",{"id":56,"doc_module":4,"doc_module_name":9,"category_name":57,"show_sort_weight":24,"slug":58},19,"General","general",{"code":4,"msg":60,"data":61},"ok",{"site_id":62,"language":63,"slug":64,"title":65,"keywords":66,"description":67,"schema_data":68,"social_meta":127,"head_meta":129,"extra_data":131,"updated_unix":133},105,"en","components-of-character-exploring-the-computational-similarity-of-austens-characters","Components of Character - Exploring the Computational Similarity of Austen’s Characters","","Understanding characters is central to computational work on literary fiction. A growing toolkit extracts character information from text, and recent research introduces AustenAlike to evaluate character-similarity representations in Jane Austen. This study extends feature-based character representations with multi-feature experiments and also derives character vectors from GPT-4-generated textual descriptions. Results show limited benefit from combining feature types, while GPT-4 description embeddings perform best on the expert benchmark, yet all representations remain difficult to match for nuanced, human-identified similarities.",{"@graph":69,"@context":126},[70,84,105],{"@type":71,"itemListElement":72},"BreadcrumbList",[73,77,79,82],{"item":74,"name":75,"@type":76,"position":8},"https://docshare.wps.com","Home","ListItem",{"item":78,"name":9,"@type":76,"position":14},"https://docshare.wps.com/document/",{"item":80,"name":40,"@type":76,"position":81},"https://docshare.wps.com/document/research-report/",3,{"item":83,"name":65,"@type":76,"position":19},"https://docshare.wps.com/document/components-of-character-exploring-the-computational-similarity-of-austens-characters/148864/",{"url":83,"name":65,"@type":85,"image":86,"author":91,"headline":65,"publisher":94,"fileFormat":97,"inLanguage":63,"description":67,"dateModified":98,"datePublished":99,"encodingFormat":97,"isAccessibleForFree":100,"interactionStatistic":101},"DigitalDocument",{"url":87,"@type":88,"width":89,"height":90},"https://docshare.wps.com/thumbnails/components-of-character-exploring-the-computational-similarity-of-austens-characters/148864.png","ImageObject",300,407,{"name":92,"@type":93},"Noah","Person",{"url":74,"name":95,"@type":96},"DocShare","Organization","application/pdf","2026-10-25","2026-08-26",true,{"@type":102,"interactionType":103,"userInteractionCount":39},"InteractionCounter",{"@type":104},"ViewAction",{"@type":106,"mainEntity":107},"FAQPage",[108,114,118,122],{"name":109,"@type":110,"acceptedAnswer":111},"What is the main goal of this paper?","Question",{"text":112,"@type":113},"To evaluate and extend computational representations of Jane Austen characters for measuring character similarity, including both feature-based and GPT-4-derived vector approaches.","Answer",{"name":115,"@type":110,"acceptedAnswer":116},"How does AustenAlike define character similarity?",{"text":117,"@type":113},"It splits similarity into three parts: narratological role, social characteristics, and wisdom-of-the-crowd comparisons extracted from historical Persuasions records.",{"name":119,"@type":110,"acceptedAnswer":120},"What does the study find about combining multiple feature types?",{"text":121,"@type":113},"Combining different feature types yields surprisingly little advantage, suggesting the features capture largely overlapping information.",{"name":123,"@type":110,"acceptedAnswer":124},"How do GPT-4-generated textual descriptions perform on the benchmark?",{"text":125,"@type":113},"Embeddings from GPT-4 descriptions align most closely with expert judgments, outperforming direct similarity elicitation, though the expert benchmark remains challenging overall.","https://schema.org",{"og:url":83,"og:type":128,"og:title":65,"og:site_name":95,"og:description":67},"article",{"robots":130,"canonical":83},"index,follow",{"doc_id":132,"site_id":62},148864,1787786956,{"code":4,"msg":5,"data":135},{"doc_id":132,"user_id":136,"nickname":92,"user_avatar":137,"doc_module":4,"category_id":39,"category_name":40,"doc_title":65,"doc_description":67,"doc_content":138,"file_id":139,"file_url":140,"file_type":141,"file_size":142,"view_count":39,"is_deleted":4,"is_public":8,"is_downloadable":8,"audit_status":8,"page_count":143,"language":144,"language_code":63,"site_id":62,"html_lang":63,"table_of_contents":145,"faqs":146,"seo_title":147,"seo_description":67,"update_tm":133,"read_time":148,"preview_page_count":4},8796095462418,"https://ap-avatar.wpscdn.com/avatar/80000253c1241d02b47?x-image-process=image/resize,m_fixed,w_180,h_180&k=1778826106357471780","Components of Character:  \nExploring the Computational Similarity of Austen’s Characters  \nCarolyn Jane Anderson 1  \n1Wellesley College, USA  \nCorresponding author: Carolyn Jane Anderson , [carolyn.anderson@wellesley.edu](carolyn.anderson@wellesley.edu)  \nAbstract  \nUnderstanding characters is a key topic of interest in computational explorations of literary fiction. There are a growing number of tools available for extracting information about characters from text. Recent work proposes AustenAlike, a three-part benchmark for Jane Austen character similarity, to evaluate computational representations of character. I extend this exploration of feature-based representations of character with new experiments, including representations build from multiple features. I also explore away of obtaining vector representations of characters using a large language model by generating short textual descriptions of characters using GPT-4, and then taking their embeddings. My results show surprisingly little advantage from combining multiple kinds of features, suggesting that the different features capture similar information to each other. I find that the LLM-generated textual descriptions perform best on expert benchmark, outperforming direct elicitation of character similarity judgments from the same model. However, the expert benchmark remains challenging across all representations explored, showing that there is much room for improvement in capturing the more nuanced aspects of character similarity that human readers identify.  \nKeywords  \nliterary fiction; characters; vector similarity  \nI INTRODUCTION  \nThere is a growing interest in using computational methods within literary analysis. Computational methods have become more accessible due to the increased availability of easy-to-run, inexpensive methods for analyzing text data. This includes both general purpose large language models, which can be used to annotate textual data [Gilardi et al., 2023, Alizadeh et al., 2025], and open-source systems designed specifically to extract information from literary fiction [Bamman et al., 2014, Yoder et al., 2021] .  \nA key area of interest is in understanding characters from literary fiction. Pipelines for analyzing literary text can be used to identify character mentions and extract information about their attributes, event roles, and speech [Bamman et al., 2014, Yoder et al., 2021] . These features can then be used to analyze characters in different ways. Previous work has sought to extract and classify character types [Chambers and Jurafsky, 2009, Bamman et al., 2014, Jahan and Finlayson, 2019, Valls-Vargas et al., 2021, Stammbach et al., 2022], build networks of relationships between characters [Elson et al., 2010, Lee and Yeung, 2012, Jayannavar et al., 2015, Agarwal et al., 2013, Wohlgenannt et al., 2016, Labatut and Bost, 2019], and study character traits [Flekova and Gurevych, 2015, Pizzolli and Strapparava, 2019, Kim and Klinger, 2019] .  \nJournal of Data Mining and Digital Humanities 1 [http://jdmdh.episciences.org](http://jdmdh.episciences.org)  \nISSN 2416-5999, an open-access journal  \nJames Morland from Northanger Abbey  \nSibling to heroine and single 20-year-old clergyman with income of £400/year Social Pairings: Charles Hayter, Edward Ferrars, Robert Martin  \nNarrative Role Pairings: Isabella Knightley, John Dashwood, Margaret Dashwood, Susan Price, William Price, Elizabeth Elliot, Mary Musgrove, Jane Bennet, Mary Bennet, Kitty Bennet, Lydia Bennet  \nExpert Pairings: Edmund Bertram, Edward Ferrars, Henry Tilney, Philip Elton  \nFigure 1: Example character from AustenAlike  \nYang and Anderson [2024] propose a benchmark that tests how well such systems extract information about characters using the works of Jane Austen as a test case. In their view, a system that produces useful character features is one whose features can be used to compare the similarity of characters. Since character similarity is a multi-faceted notion, their A","cbCaiqX00fOXbgp2","https://ap.wps.com/l/cbCaiqX00fOXbgp2","pdf",453225,26,"English","# Abstract\n# Introduction\n## Computational interest in literary analysis\n## Character extraction and representation\n## AustenAlike benchmark and similarity views\n## This paper’s extensions and methods","[{\"question\":\"What is the main goal of this paper?\",\"answer\":\"To evaluate and extend computational representations of Jane Austen characters for measuring character similarity, including both feature-based and GPT-4-derived vector approaches.\"},{\"question\":\"How does AustenAlike define character similarity?\",\"answer\":\"It splits similarity into three parts: narratological role, social characteristics, and wisdom-of-the-crowd comparisons extracted from historical Persuasions records.\"},{\"question\":\"What does the study find about combining multiple feature types?\",\"answer\":\"Combining different feature types yields surprisingly little advantage, suggesting the features capture largely overlapping information.\"},{\"question\":\"How do GPT-4-generated textual descriptions perform on the benchmark?\",\"answer\":\"Embeddings from GPT-4 descriptions align most closely with expert judgments, outperforming direct similarity elicitation, though the expert benchmark remains challenging overall.\"}]","Components of Character - Exploring the Computational Similarity of Austen’s Characters | PDF",66]