@prefix this: . @prefix sub: . @prefix np: . @prefix dct: . @prefix xsd: . @prefix rdfs: . @prefix prov: . @prefix npx: . sub:Head { this: np:hasAssertion sub:assertion; np:hasProvenance sub:provenance; np:hasPublicationInfo sub:pubinfo; a np:Nanopublication . } sub:assertion { dct:title "AI Agent-Driven Framework for Automated Product Knowledge Graph Construction in E-Commerce"; ; , , , , , , , , , , ; a prov:Entity . dct:subject ; a ; rdfs:comment "This framework uses LLM-powered agents to automate the entire process of Knowledge Graph construction. It involves three main stages: ontology creation and expansion (entity/relation discovery), ontology refinement (schema improvement), and knowledge graph population (instance extraction) from unstructured product descriptions, directly addressing KG construction tasks."; rdfs:label "AI Agent-Driven Framework"; . a ; rdfs:comment "Presented by Bi et al., CodeKGC is a method that employs code LLMs and schema-aware prompts for generative KG tasks, enhanced by rationale-based generation."; rdfs:label "CodeKGC" . a ; rdfs:comment "Proposed by Cui et al., KG-ICL is a method that combines in-context learning with a prompt graph and unified tokenization to achieve strong generalization across various KG tasks."; rdfs:label "KG-ICL" . a ; rdfs:comment "Grapher is discussed as a method that generates end-to-end KGs by dividing tasks into entity and relation extraction using pre-trained Language Models."; rdfs:label "Grapher" . a ; rdfs:comment "This method is mentioned as a prior LLM-based pipeline that aims to minimize human intervention in KG construction by incorporating competency-question-based ontology generation and LLM-driven evaluation."; rdfs:label "Kommineni et al.'s Pipeline" . a ; rdfs:comment "Proposed by Lairgi et al., this method is presented as a zero-shot, plug-and-play system for distilling unstructured input into KGs with minimal post-processing."; rdfs:label "iText2KG" . a ; rdfs:comment "This method by Ning et al. explores factual knowledge extraction for KGs via prompt templates and parameter tuning to improve accuracy."; rdfs:label "Ning et al.'s Factual Knowledge Extraction" . a ; rdfs:comment "Xie et al. introduced PromptKG, a toolkit designed to integrate prompt-learning methods for various Knowledge Graph applications."; rdfs:label "PromptKG" . a ; rdfs:comment "This embedding method, proposed by Xu et al., is tailored for product KGs and uses multimodal data to enhance KG completion and recommendation tasks."; rdfs:label "Xu et al.'s Embedding Method" . a ; rdfs:comment "This method, presented by Yao et al., utilizes entity/relation prompts for triple classification and relation prediction, focusing on Knowledge Graph completion tasks."; rdfs:label "KG-LLM" . a ; rdfs:comment "This method, proposed by Zhang et al., is discussed as a prior LLM-based approach for automated KG construction, which separates open information extraction, schema creation, and canonicalization."; rdfs:label "EDC Framework" . a ; rdfs:comment "This framework, introduced by Zhu et al., is discussed as a multi-agent system that combines LLMs with external data for KG construction."; rdfs:label "AutoKG Framework" . } sub:provenance { sub:assertion prov:wasAttributedTo ; prov:wasDerivedFrom . } sub:pubinfo { this: dct:created "2026-03-13T16:02:03.618Z"^^xsd:dateTime; dct:creator ; npx:hasNanopubType ; npx:supersedes ; rdfs:label "LLM-KG assessment for paper 10.48550/arXiv.2511.11017" . sub:sig npx:hasAlgorithm "RSA"; npx:hasPublicKey "MIIBIjANBgkqhkiG9w0BAQEFAAOCAQ8AMIIBCgKCAQEAwNz2QK3SEifno78S7+48zUB0xpTex3mAzW73ZimHqNcdEMU5/apslrGrTHGFAt/Chocgo++r6JQp5ygY7NyJHGWdaIqnt85pjX4PbNfLAvapyUO00qZP34fY61w4eZ9UMtleWEsmZKRtQPyJ8ODl46i/rfPuZlcJGpM9Nmy5mpGWuepqIEvF4a/t7pLVeCEDFSYXT+yaiygt6ynIK5f7TtEDhZpeUf/Q74WhMPJXm4yTU/hqOX4IW+50kWHNArGGZwUaXwzyG6M3Zd6UMModryGkLqS4H/MSE3ZA1Ylnms7BfWLEXhMWlaKi6HRV4nGRDLhxVSi9LSRi3LWKLhNIIQIDAQAB"; npx:hasSignature "iLSElcgqqBsWTLtic7yrdoX2HgsuT/lKI7xDXW5jloh4uxRGPdY72/D+LdNE8CNXSiWOI5ndpSWJhachzVbWHxLqG/f1ihRAQTBOQrQ2U49UvOZN5SXBLqMzk0inB3uXMduZBmiSk9EhOpn+KY/DNcZOHnMtpE0gXLkOxmVdvtIlpQzr2QsLAdj02YMEC01ndKNcmyXL8wvcUrTnmHreZjxjeaAzElGlCmRV0+yVNisxrrgm0Ur+Y5YwDmxMtRob2HFlJkjc1l/mkeMOe5IRHEp87HyMppCQhl0WW4lR9XkHfdbYrafIummDmzfrvg1R+JnTnILVuC7HVin5rFAswA=="; npx:hasSignatureTarget this:; npx:signedBy . }