Animer une base de connaissance: des ontologies aux mod{è}les d'I.A. g{é}n{é}rative

• URL: http://arxiv.org/abs/2509.01304v1
• Date: Mon, 01 Sep 2025 09:40:55 GMT
• Title: Animer une base de connaissance: des ontologies aux mod{è}les d'I.A. g{é}n{é}rative
• Authors: Peter Stockinger,
• Abstract summary: Article proposes a reading of the hybridization between symbolic AI and neural (or sub-symbolic) AI based on a field of application.<n>We describe the LaCAS ecosystem -- Open Archives in Linguistic and Cultural Studies.<n>We illustrate this approach using the knowledge domain ''Languages of the world'' (540 languages) and the knowledge object ''Quechua (language)''
• Score: 0.0
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: In a context where the social sciences and humanities are experimenting with non-anthropocentric analytical frames, this article proposes a semiotic (structural) reading of the hybridization between symbolic AI and neural (or sub-symbolic) AI based on a field of application: the design and use of a knowledge base for area studies. We describe the LaCAS ecosystem -- Open Archives in Linguistic and Cultural Studies (thesaurus; RDF/OWL ontology; LOD services; harvesting; expertise; publication), deployed at Inalco (National Institute for Oriental Languages and Civilizations) in Paris with the Okapi (Open Knowledge and Annotation Interface) software environment from Ina (National Audiovisual Institute), which now has around 160,000 documentary resources and ten knowledge macro-domains grouping together several thousand knowledge objects. We illustrate this approach using the knowledge domain ''Languages of the world'' (~540 languages) and the knowledge object ''Quechua (language)''. On this basis, we discuss the controlled integration of neural tools, more specifically generative tools, into the life cycle of a knowledge base: assistance with data localization/qualification, index extraction and aggregation, property suggestion and testing, dynamic file generation, and engineering of contextualized prompts (generic, contextual, explanatory, adjustment, procedural) aligned with a domain ontology. We outline an ecosystem of specialized agents capable of animating the database while respecting its symbolic constraints, by articulating model-driven and data-driven methods.

Related papers

• From Text to Network: Constructing a Knowledge Graph of Taiwan-Based China Studies Using Generative AI [0.0]
  Taiwanese China Studies (CS) has developed into a rich, interdisciplinary research field shaped by the unique geopolitical position and long standing academic engagement with Mainland China.<n>This study proposes an AI assisted approach that transforms unstructured academic texts into structured, interactive knowledge representations.
  arXiv  Detail & Related papers  (2025-05-15T08:51:53Z)
• Symbolic Learning Enables Self-Evolving Agents [55.625275970720374]
  We introduce agent symbolic learning, a systematic framework that enables language agents to optimize themselves on their own. Agent symbolic learning is designed to optimize the symbolic network within language agents by mimicking two fundamental algorithms in connectionist learning. We conduct proof-of-concept experiments on both standard benchmarks and complex real-world tasks.
  arXiv  Detail & Related papers  (2024-06-26T17:59:18Z)
• Sanskrit Knowledge-based Systems: Annotation and Computational Tools [0.12086712057375555]
  We address the challenges and opportunities in the development of knowledge systems for Sanskrit. This research contributes to the preservation, understanding, and utilization of the rich linguistic information embodied in Sanskrit texts.
  arXiv  Detail & Related papers  (2024-06-26T12:00:10Z)
• DIVKNOWQA: Assessing the Reasoning Ability of LLMs via Open-Domain Question Answering over Knowledge Base and Text [73.68051228972024]
  Large Language Models (LLMs) have exhibited impressive generation capabilities, but they suffer from hallucinations when relying on their internal knowledge. Retrieval-augmented LLMs have emerged as a potential solution to ground LLMs in external knowledge.
  arXiv  Detail & Related papers  (2023-10-31T04:37:57Z)
• Semantic Similarity Measure of Natural Language Text through Machine Learning and a Keyword-Aware Cross-Encoder-Ranking Summarizer -- A Case Study Using UCGIS GIS&T Body of Knowledge [2.4909170697740968]
  GIS&T Body of Knowledge (BoK) is a community-driven endeavor to define, develop, and document geospatial topics. This research evaluates the effectiveness of multiple natural language processing (NLP) techniques in extracting semantics from text. It also offers a new perspective on the use of machine learning techniques for analyzing scientific publications.
  arXiv  Detail & Related papers  (2023-05-17T01:17:57Z)
• O-Dang! The Ontology of Dangerous Speech Messages [53.15616413153125]
  We present O-Dang!: The Ontology of Dangerous Speech Messages, a systematic and interoperable Knowledge Graph (KG) O-Dang! is designed to gather and organize Italian datasets into a structured KG, according to the principles shared within the Linguistic Linked Open Data community. It provides a model for encoding both gold standard and single-annotator labels in the KG.
  arXiv  Detail & Related papers  (2022-07-13T11:50:05Z)
• Taxonomy Enrichment with Text and Graph Vector Representations [61.814256012166794]
  We address the problem of taxonomy enrichment which aims at adding new words to the existing taxonomy. We present a new method that allows achieving high results on this task with little effort. We achieve state-of-the-art results across different datasets and provide an in-depth error analysis of mistakes.
  arXiv  Detail & Related papers  (2022-01-21T09:01:12Z)
• Knowledge Graph Augmented Network Towards Multiview Representation Learning for Aspect-based Sentiment Analysis [96.53859361560505]
  We propose a knowledge graph augmented network (KGAN) to incorporate external knowledge with explicitly syntactic and contextual information. KGAN captures the sentiment feature representations from multiple perspectives, i.e., context-, syntax- and knowledge-based. Experiments on three popular ABSA benchmarks demonstrate the effectiveness and robustness of our KGAN.
  arXiv  Detail & Related papers  (2022-01-13T08:25:53Z)
• Human Activity Recognition Models in Ontology Networks [0.5801044612920815]
  Arianna+ is a framework to design networks of for representing knowledge enabling smart homes to perform recognition online. Arianna+ provides a flexible interface between the inputs and outputs of procedures and statements. Arianna+ schedules procedures on the basis of events by employing logic-based reasoning.
  arXiv  Detail & Related papers  (2021-05-05T18:23:56Z)
• A New Neural Search and Insights Platform for Navigating and Organizing AI Research [56.65232007953311]
  We introduce a new platform, AI Research Navigator, that combines classical keyword search with neural retrieval to discover and organize relevant literature. We give an overview of the overall architecture of the system and of the components for document analysis, question answering, search, analytics, expert search, and recommendations.
  arXiv  Detail & Related papers  (2020-10-30T19:12:25Z)
• Machine Knowledge: Creation and Curation of Comprehensive Knowledge Bases [28.856786775318486]
  Large-scale knowledge bases, also known as knowledge graphs, have been automatically constructed from web contents and text sources. This article surveys fundamental concepts and practical methods for creating and large knowledge bases.
  arXiv  Detail & Related papers  (2020-09-24T09:28:13Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.