Reinforcement learning based deep fuzzy hierarchical clustering to generate personalized non-fungible token artwork

In the realm of information technology and software development, digital assets increasingly manifest as Non-Fungible Tokens (NFTs) on blockchain platforms, embodying intrinsic material value that enhances user satisfaction. Despite their potential, the creation of NFTs remains costly, time-consumin...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Neurocomputing (Amsterdam) Jg. 659; S. 131821
Hauptverfasser: Daliri, Arman, Mahdavi, Nora, Zabihimayvan, Mahdieh, Norouzi Baranghar, Aynaz, Zaeimzadeh, Nima, mohammadzadeh, Javad
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Elsevier B.V 01.01.2026
Schlagworte:
Clustering
Deep learning
Digital Arts
Fuzzy
Hierarchical
Non-Fungible Token Artwork
Non-Fungible Token Artwork
Hierarchical
Deep learning
Fuzzy
Clustering
Digital Arts
ISSN:0925-2312
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:In the realm of information technology and software development, digital assets increasingly manifest as Non-Fungible Tokens (NFTs) on blockchain platforms, embodying intrinsic material value that enhances user satisfaction. Despite their potential, the creation of NFTs remains costly, time-consuming, and labor-intensive. To address these challenges, we introduce a digital asset generation engine specifically designed for producing NFT artwork. Utilizing real-world datasets curated by digital art experts, our engine synthesizes individual image layers into cohesive, high-quality artistic outputs. Leveraging artificial intelligence, we employ a novel Deep Fuzzy Hierarchical Clustering approach, which integrates autoencoder neural networks, fuzzy clustering, and hierarchical clustering methods. This integrated approach enables precise classification of image layers, achieving an impressive accuracy rate of 95 %. Here, we demonstrate the potential of AI-enhanced solutions in the digital art and NFT space. Our engine not only reduces costs and labor intensity in digital art production but also allows users to personalize their NFT collections by selecting desired layers and specifying rarity, arrangement order, and metadata details. This study underscores the significance of intersectional research between artificial intelligence and fine arts, opening avenues for future advancements in computational art analysis and creative AI applications. •Proposing a new method for personalizing NFT using layered image structures.•Utilizing real-world art layers, designed by experts, to train and test the system.•Addressing the automatic selection of clusteres using the WHH , ensuring efficient NFT layer categorization.•Introducing a Deep Fuzzy Hierarchical Clustering using autoencoders, fuzzy-cluster, and hierarchical-cluster for image classification.•Enabling users to control layer selection, arrangement, and metadata output.
ISSN:0925-2312
DOI:10.1016/j.neucom.2025.131821