[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-0-en-105":3,"doc-seo-292967-105":59,"doc-detail-292967-en":131},{"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":124,"head_meta":126,"extra_data":128,"updated_unix":130},105,"en","predicting-mbti-personalities-of-tv-show-characters-research-paper","Predicting MBTI Personalities of TV Show Characters - Research Paper","","This paper proposes a multi-task BERT approach to predict Myers-Briggs Type Indicator (MBTI) personality types from dialogue text in TV shows. The model infers personality from a sliding window of consecutive utterances spoken by a character. To avoid overfitting to character identities, character identification is included as an auxiliary task with gradient reversal, and the training loss emphasizes personality prediction. Separate models are trained for each MBTI dimension and evaluated on 8 unseen characters, outperforming baselines.",{"@graph":69,"@context":123},[70,84,106],{"@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/predicting-mbti-personalities-of-tv-show-characters-research-paper/292967/",{"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/predicting-mbti-personalities-of-tv-show-characters-research-paper/292967.png","ImageObject",300,407,{"name":92,"@type":93},"Taylor Morgan","Person",{"url":74,"name":95,"@type":96},"DocShare","Organization","application/pdf","2026-11-10","2026-09-17",true,{"@type":102,"interactionType":103,"userInteractionCount":105},"InteractionCounter",{"@type":104},"ViewAction",18,{"@type":107,"mainEntity":108},"FAQPage",[109,115,119],{"name":110,"@type":111,"acceptedAnswer":112},"What is the main goal of the proposed model?","Question",{"text":113,"@type":114},"The model aims to predict MBTI personality types from TV show dialogues by using a window of consecutive lines spoken by each character.","Answer",{"name":116,"@type":111,"acceptedAnswer":117},"How does the paper reduce the risk of memorizing character identities?",{"text":118,"@type":114},"It adds character identification as a second task and applies gradient reversal so the model is discouraged from learning identity-specific signals.",{"name":120,"@type":111,"acceptedAnswer":121},"How is the model evaluated?",{"text":122,"@type":114},"The authors evaluate on a dataset of utterances from 8 unseen characters and report results that outperform a vanilla BERT baseline and prior models.","https://schema.org",{"og:url":83,"og:type":125,"og:title":65,"og:site_name":95,"og:description":67},"article",{"robots":127,"canonical":83},"index,follow",{"doc_id":129,"site_id":62},292967,1789658118,{"code":4,"msg":5,"data":132},{"doc_id":129,"user_id":133,"nickname":92,"user_avatar":134,"doc_module":4,"category_id":39,"category_name":40,"doc_title":65,"doc_description":67,"doc_content":135,"file_id":136,"file_url":137,"file_type":138,"file_size":139,"view_count":105,"is_deleted":4,"is_public":8,"is_downloadable":8,"audit_status":8,"page_count":24,"language":140,"language_code":63,"site_id":62,"html_lang":63,"table_of_contents":141,"faqs":142,"seo_title":143,"seo_description":67,"update_tm":130,"read_time":144,"preview_page_count":4},1099523885336,"https://ap-avatar.wpscdn.com/davatar_276721f389ce27ea32af1340a28f341c","Predicting MBTI personalities of TV show characters  \nGoutham Karunakaran  \nCharles University [goutham794@gmail.com](goutham794@gmail.com)  \nRudolf Rosa  \nCharles University [rosa@ufal.mff.cuni.cz](rosa@ufal.mff.cuni.cz)  \nAbstract  \nThis paper proposes a novel multi-task BERT model for predicting Myers-Briggs Type Indicator (MBTI) personality types from dialoguesin TV shows. The model is designed to predict the personality type of a character based on a window of consecutive lines spoken by that character. In this paper, \"character\" refers to the person in the TV show who utters the lines and \"personality\" to the MBTI traits. The baseline approach tends to memorize identities of individual characters instead of learning features of personality traits. Therefore, we add character identification as a second task with gradient reversal to force the model not to learn that task. For the loss, a larger weight is given to personality loss over character identification loss. Separate models are trained for each dimension of MBTI. After training, we evaluated the model on a dataset consisting of utterances of 8 unseen characters. The results show that our proposed model outperforms our baseline with vanilla BERT and previous models trained on users’ social media posts. Our findings demonstrate the potential of gradient reversal for improving personality prediction from utterances.  \n1 Introduction  \nPersonality prediction from text is a challenging task in natural language processing (NLP) . The accurate prediction of personality traits becomes useful when considering generated dialogues and scripts. To create good narratives, the lines generated for each character should align with their persona and be consistent with their established personality traits. One such work by (Rosa et al., 2022) is about generating coherent theatre plays by adapting large language models.  \nOne widely-used model for personality prediction is the Myers-Briggs Type Indicator (MBTI), which classifies individuals into one of 16 personality types based on four dichotomous dimensions:  \nIntroversion vs. Extraversion (I-E), Sensing vs. Intuition (S-N), Thinking vs. Feeling (T-F), and Judging vs. Perceiving (J-P) . While predicting MBTI from text has been attempted in previous works, it remains a difficult task due to the complexity and variability of natural language expressions. MBTI has been subject to criticism and skepticism from various quarters. Some researchers have raised questions about the scientific validity and reliability of MBTI, arguing that it lacks empirical support and does not meet the standards of psychometric testing (Pittenger, 2005) . However, data with MBTI ratings are easily available, which makes MBTI classification a viable NLP task.  \nIn this paper, we propose a novel multi-task BERT model for predicting MBTI personality types from dialogues in TV shows. We chose TV show dialogues as our dataset because they provide a rich source of natural language expressionsand character interactions, which are expected to reveal personality traits. One of the challenges in predicting MBTI from text is that personality traits are not always clearly reflected in the surface-level linguistic features. Moreover, the expression of personality traits can vary greatly depending on context and the individual’s communication style.  \n2 Related Work  \nAutomatic personality detection from text and social media data has received considerable attention in recent years. Personality trait prediction used tobe done using self-reported surveys and questionnaires which were limited by small sample sizesand self-report biases. Plank and Hovy (2015) developed predictive models using linguistic features and user metadata on a large corpus of tweets with self-reported MBTI types of the users. Their models showed slightly better performance than the majority baseline, but only for the I-E and T-F dimensions.  \nA contemporary work by Sang et al. (2022) also  \ntries to","cbCaiacN7FzUCVbq","https://ap.wps.com/l/cbCaiacN7FzUCVbq","pdf",464007,"English","# Abstract\n# Introduction\n# Related Work\n# Data\n# Methodology\n# Experiments and Results","[{\"question\":\"What is the main goal of the proposed model?\",\"answer\":\"The model aims to predict MBTI personality types from TV show dialogues by using a window of consecutive lines spoken by each character.\"},{\"question\":\"How does the paper reduce the risk of memorizing character identities?\",\"answer\":\"It adds character identification as a second task and applies gradient reversal so the model is discouraged from learning identity-specific signals.\"},{\"question\":\"How is the model evaluated?\",\"answer\":\"The authors evaluate on a dataset of utterances from 8 unseen characters and report results that outperform a vanilla BERT baseline and prior models.\"}]","Predicting MBTI Personalities of TV Show Characters - Research Paper | PDF",13]