[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-83874-en":3,"doc-seo-83874-105":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":20,"is_deleted":4,"is_public":21,"is_downloadable":21,"audit_status":21,"page_count":22,"language":23,"language_code":24,"site_id":25,"html_lang":24,"table_of_contents":26,"faqs":27,"seo_title":13,"seo_description":14,"update_tm":28,"read_time":29},83874,8796095461564,"Liam","https://ap-avatar.wpscdn.com/davatar_155a257f0dc6eb9ab79c44ca47cae57d",8,"Research & Report","LLM for the Development of FCM","The article presents a method for developing a fuzzy cognitive map (FCM) using a local large language model (LLM) to extract quantitative signals from text. A local model such as Qwen2.5-32B ingests entity concepts as prompts and outputs relevant quantities, which are then used to construct a data-driven FCM. The approach is implemented and thoroughly tested with TripAdvisor hotel reviews, followed by training and evaluation of an unfiltered document-derived FCM.","arXiv :2607 .04983v2 [ cs .NE] 9 Jul 2026  \nLLM for the development of FCM  \nAlexis Kafantaris  \nJune 2026  \nAbstract  \nThis article is about the development of a fuzzy cognitive map using a local large language model. In light of recent advances, it is evident that large language models, and even local large language models, are capable of extracting quantities from textual data. In other words, a local LLM like Qwen2.5-32B, or probably larger, can accept entities as prompt input and determine relevant quantitative data as the model output. In turn, this output can be utilized for the construction of a data driven fuzzy cognitive map. Hence, this implementation is achieved, and then the model is thoroughly tested; Qwen2.5-32B is used and the data is extracted from hotel reviews from TripAdvisor. Furthermore, the extracted documents pass through the model unfiltered and then a fuzzy cognitive map is trained and evaluated. A case is made about Greek reviews where a star topology FCM is formed that indicates the preferences of the reviewers. Finally, external validation is performed to establish whether the fuzzy cognitive map can correlate the star rating of thereview—an outcome outside the model’s inference scope—with its predicted satisfaction.  \n1 Introduction  \n1.1 LLM  \nOne of the most recent advances in the scientific field is the large language model (LLM) technology. The LLM technology has redefined many industries, and the LLM integration helps automate various tasks. These tasks range from data collection, to data generation for personalized reviews [1], to service design process planning [2] . Another interesting use case for LLMs is the identification of topics or causal relationships. LLMs can understand context and identify topics from key words better than traditional models such as latent Dirichlet allocation [5, 6] . Hence, assisting implicitly in improving services.  \nAn auto-regressive large language model refers to a model, Qwen2.5-32B, which uses a prediction equation as follows  \nn  \nLCLM = − Xlog P(xi | x 1 ,..., xi−1;Θ)  \ni=1  \nFurthermore, except for topic modeling, LLM can also identify causality; one would need a very large LLM for that, however, there is another way that is possible to create a fuzzy cognitive map (FCM) . So, there are several implementations of an FCM, one explicitly from the father of FCM Bart Kosko that exploit specifically LLMs [15] . Kosko’s idea was to automate the pipeline to generate an FCM from nouns and then connect the entities that are derived and determine causal relationships. A similar idea has also been implemented, while the LLM extracted causal relationships from various entities that were given [7] .  \nAt this point a gap is identified, that has to do with the use of the LLM for the development of an FCM. Instead of relying on the LLM to extract causality [15, 17], the LLM can extract data from which an FCM can be trained. The user provides entity concepts, the LLM extracts the data, and then from the extracted elements a data driven gradient descent based FCM is formed. This might seem obvious at first, and it is a simple idea. To our knowledge, no prior work has used an LLM to extract data for the development of a data driven FCM.  \n1.2 FCM  \nA fuzzy cognitive map is a signed digraph; a symbolic network with concepts as nodes and fuzzy relationships as edges. By using fuzzy relations as edges it becomes interpretable, e.g. concept C1 affects concept C2 by 0.2, edge C1C2 is 0.2, or concept C3 affects concept C2 by negative 0.5, edge C3C2 is-0.5. The FCM is a shallow recurrent neural network that, instead of calculating random weights, calculates weights as system dynamics [8] . In this way, one can simulate the system and, through simulation, determine the mathematical equilibrium [9] .  \nFuzzy cognitive map learning methods are an interesting topic that has been thoroughly researched[10] . Some methods have convergence advantages, while others have interpretabili","cbCainj1NS2BHw52","https://ap.wps.com/l/cbCainj1NS2BHw52","pdf",306511,3,1,14,"English","en",105,"# Introduction\n## LLM\n## FCM\n# Background","[{\"question\":\"How does the method use a local LLM to develop a fuzzy cognitive map?\",\"answer\":\"The user provides entity concepts as prompts, the local LLM extracts relevant quantities from the text, and those extracted elements are used to train a data-driven FCM.\"},{\"question\":\"What data source and model are used in the experiments?\",\"answer\":\"Experiments use hotel reviews from TripAdvisor and the Qwen2.5-32B local LLM to extract data and build the fuzzy cognitive map.\"},{\"question\":\"How is the fuzzy cognitive map validated in the study?\",\"answer\":\"An external validation checks whether the trained fuzzy cognitive map correlates with the star rating outcomes and predicts reviewer satisfaction beyond the model’s direct inference scope.\"}]",1784191142,35,{"code":4,"msg":31,"data":32},"ok",{"site_id":25,"language":24,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"llm-for-the-development-of-fcm","",{"@graph":36,"@context":85},[37,53,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,50],{"item":41,"name":42,"@type":43,"position":21},"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":20},"https://docshare.wps.com/document/research-report/",{"item":51,"name":13,"@type":43,"position":52},"https://docshare.wps.com/document/llm-for-the-development-of-fcm/83874/",4,{"url":51,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":24,"description":14,"dateModified":61,"datePublished":62,"encodingFormat":60,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":41,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-07-24","2026-07-16",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"How does the method use a local LLM to develop a fuzzy cognitive map?","Question",{"text":75,"@type":76},"The user provides entity concepts as prompts, the local LLM extracts relevant quantities from the text, and those extracted elements are used to train a data-driven FCM.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"What data source and model are used in the experiments?",{"text":80,"@type":76},"Experiments use hotel reviews from TripAdvisor and the Qwen2.5-32B local LLM to extract data and build the fuzzy cognitive map.",{"name":82,"@type":73,"acceptedAnswer":83},"How is the fuzzy cognitive map validated in the study?",{"text":84,"@type":76},"An external validation checks whether the trained fuzzy cognitive map correlates with the star rating outcomes and predicts reviewer satisfaction beyond the model’s direct inference scope.","https://schema.org",{"og:url":51,"og:type":87,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":89,"canonical":51},"index,follow",{"doc_id":7,"site_id":25},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"id":21,"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":52,"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":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]