[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-81698-en":3,"doc-seo-81698-105":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},81698,3848291630094,"Emma Wilson","https://eur-avatar.wpscdn.com/davatar_085a072bc5b1113ac321206ff7593b45",8,"Research & Report","From Strings to Things for Personal Knowledge Graphs Evaluating LLM Triple Extraction for Recommendation Systems","Personal Knowledge Graphs (PKGs) provide a privacy-preserving way to represent user preferences, yet building them from unstructured, decentralized conversation data remains difficult. The work proposes a reproducible pipeline that converts conversational “strings” into semantic “things” by extracting structured user-preference triples with lightweight LLMs. It evaluates Qwen- and Gemma-based models for RDF-compliant triple extraction linked to Wikidata identifiers. The study measures both semantic fidelity and downstream recommendation usefulness, finding models whose triple extraction quality aligns with stronger recommendation performance.","From “Strings” to “Things” for Personal Knowledge Graphs: Evaluating LLM Triple Extraction for Recommendation Systems  \nAbhirup Dasgupta*,†, Fernando Spadea*,† and Oshani Seneviratne* Rensselaer Polytechnic Institute, Troy NY 12180, USA  \nAbstract  \nPersonal Knowledge Graphs (PKGs) offer a privacy-preserving framework for modeling user preferences, yet constructing them from unstructured, decentralized conversational data remains a challenge. This paper bridges the gap between conversational “strings” and semantic “things”by presenting a reproducible pipeline for extracting structured user-preference triples using lightweight Large Language Models (LLMs) . We evaluate Qwen-and Gemma-based models on their ability to extract RDF-compliant triples linked to Wikidata identifiers from conversational data for PKG construction. Our evaluation assesses both the semantic extraction fidelity and the utility of the resulting graphs in a downstream recommendation task. We found that certain models performed well and had proportionally high downstream performance relative to their triple extraction performance.  \nKeywords  \nPersonal Knowledge Graph Construction, Large Language Models, Triple Extraction, Conversational Recommendation Systems, Personalized AI, Decentralized Personalization  \n1. Introduction  \nPersonalized Knowledge Graphs (PKGs) have emerged as a powerful paradigm for representing user-specific preferences, contextual signals, and behavioral patterns in a structured and machineinterpretable form. Unlike global knowledge graphs that encode only widely “notable” facts, PKGs capture fine-grained, transient, and often non-canonical personal information essential for personalization, user modeling, and adaptive AI systems. Foundational work by Balog and Kenter [1] formally articulated the promise and challenges of PKGs, while subsequent surveys [2, 3] have underscored the need for scalable construction pipelines and interoperable representations. Complementary advancesin KG-enhanced recommendation [4] and conversational preference elicitation [5] further demonstrate that explicitly structured relational signals consistently improve personalization quality.  \nBuilding on this trajectory, we investigate a dialogue-driven PKG construction approach that leverages open-weight large language models (LLMs) to extract user–item relational triples, validate their semantic fidelity, and assess their downstream utility in recommendation settings. In doing so, this work advances creating, managing, and exploiting personalized, dynamically generated KGs that remain interpretable, interoperable, and semantically grounded.  \nAlthough pre-trained LLMs excel at open-domain knowledge extraction, their capacity to construct PKGs, particularly from conversational interactions, remains underexplored. Traditional recommender systems that incorporate structured user–item relationships via graph-based neural models have demonstrated clear gains in accuracy and interpretability [6, 7] . However, such methods typically depend on centrally logged, explicit interaction traces, which are often unavailable due to privacy concerns or infrastructural constraints. Privacy considerations may restrict the collection or sharing of detailed behavioral data, while infrastructural limitations may prevent platforms from maintaining persistent user identifiers, centralized logging pipelines, or cross-service data aggregation. This motivates an  \nLLM-Text2KG’26: 5th International Workshop on LLM-Integrated Knowledge Graph Generation from Text (Co-located with ESWC 2026), May 10–14, 2026, Dubrovnik, Croatia  \n* Corresponding author.  \n†  \nThese authors contributed equally.  \n􀀚 0009-0002-8434-9754 (A. Dasgupta); 0009-0006-4278-3666 (F. Spadea); 0000-0001-8518-917X (O. Seneviratne)  \n © 2026 Copyright for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0) .  \nalternative paradigm: extracting user preferences dir","cbCaiasn32Ire683","https://ap.wps.com/l/cbCaiasn32Ire683","pdf",856435,3,1,16,"English","en",105,"# Introduction\n## Motivation for Decentralized PKG Construction\n## LLM-Driven Triple Extraction Pipeline\n## Evaluation Setup and Dataset\n## Contributions","[{\"question\":\"What problem does the paper address in constructing personal knowledge graphs?\",\"answer\":\"It targets the challenge of building PKGs from unstructured, decentralized conversational data while preserving interpretability and privacy. The goal is to transform dialogue-derived information into structured preference knowledge graphs.\"},{\"question\":\"How does the proposed approach extract knowledge from conversations?\",\"answer\":\"It uses lightweight large language models to extract (User, Relation, Item) triples from multi-turn recommendation dialogues. The triples are made RDF-compliant and linked to Wikidata identifiers using an explicit semantic schema with IRIs.\"},{\"question\":\"What aspects are evaluated to judge the quality of LLM triple extraction?\",\"answer\":\"The evaluation considers semantic extraction fidelity and the utility of the resulting graphs in a downstream recommendation task. The paper also examines how extraction performance relates to recommendation effectiveness.\"}]",1784175479,40,{"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},"from-strings-to-things-for-personal-knowledge-graphs-evaluating-llm-triple-extraction-for-recommendation-systems","",{"@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/from-strings-to-things-for-personal-knowledge-graphs-evaluating-llm-triple-extraction-for-recommendation-systems/81698/",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","2026-07-22","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},"What problem does the paper address in constructing personal knowledge graphs?","Question",{"text":75,"@type":76},"It targets the challenge of building PKGs from unstructured, decentralized conversational data while preserving interpretability and privacy. The goal is to transform dialogue-derived information into structured preference knowledge graphs.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does the proposed approach extract knowledge from conversations?",{"text":80,"@type":76},"It uses lightweight large language models to extract (User, Relation, Item) triples from multi-turn recommendation dialogues. The triples are made RDF-compliant and linked to Wikidata identifiers using an explicit semantic schema with IRIs.",{"name":82,"@type":73,"acceptedAnswer":83},"What aspects are evaluated to judge the quality of LLM triple extraction?",{"text":84,"@type":76},"The evaluation considers semantic extraction fidelity and the utility of the resulting graphs in a downstream recommendation task. The paper also examines how extraction performance relates to recommendation effectiveness.","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},[93,97,101,105,110,115,119,122,127,130,134],{"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":29,"slug":118},7,"Healthcare","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"]