Human Versus Machine Intelligence: Assessing Natural Language Generation Models Through Complex Systems Theory

Enrico De Santis, Alessio Martino, Antonello Rizzi
IEEE Transactions on Pattern Analysis and Machine Intelligence, 46(7), 4812–4829, 2024.

Complex systems analysis of human- and machine-generated language through multifractality, recurrence, quantitative linguistics, and machine learning.

Complexity Science · Generative AI
DOI · IRIS

Multifractal Characterization of Texts for Pattern Recognition: On the Complexity of Morphological Structures in Modern and Ancient Languages

Enrico De Santis, Giovanni De Santis, Antonello Rizzi
IEEE Transactions on Pattern Analysis and Machine Intelligence, 45(8), 10143–10160, 2023.

Multifractal and long-range analysis of linguistic structures, connecting complex systems science with Pattern Recognition.

Language · Multifractality
DOI · IRIS

Prototype Theory Meets Word Embedding: A Novel Approach for Text Categorization via Granular Computing

Enrico De Santis, Antonello Rizzi
Cognitive Computation, 15, 976–997, 2023.

An integration of Prototype Theory, Conceptual Spaces, Granular Computing, and distributed language representations.

Representation · Granular Computing
DOI · IRIS

Modeling Failures in Smart Grids by a Bilinear Logistic Regression Approach

Enrico De Santis, Antonello Rizzi
Neural Networks, 174, 106245, 2024.

An interpretable approach to fault modelling in power grids, linking machine learning, explainability, and complex engineering systems.

Explainable AI · Smart Grids
DOI · IRIS

Degradation Mechanisms and Differential Curve Modeling for Non-Invasive Diagnostics of Lithium Cells: An Overview

Enrico De Santis, Vanessa Pennazzi, Massimiliano Luzi, Antonello Rizzi
Renewable and Sustainable Energy Reviews, 211, 115349, 2025.

A systematic framework for lithium-ion battery degradation and non-invasive diagnostics supporting data-driven health assessment.

Energy Systems · Battery Diagnostics
DOI · IRIS

From Bag-of-Words to Transformers: A Comparative Study for Text Classification in Healthcare Discussions in Social Media

Enrico De Santis, Alessio Martino, Francesca Ronci, Antonello Rizzi
IEEE Transactions on Emerging Topics in Computational Intelligence, 9(1), 1063–1077, 2025.

Compares classical text representations, BERT-based architectures, and contemporary LLMs, examining the evolution of semantic representation in a real-world NLP task.

Transformers · Text Representation
DOI · IRIS

Modeling and Recognition of Smart Grid Faults by a Combined Approach of Dissimilarity Learning and One-Class Classification

Enrico De Santis, Lorenzo Livi, Alireza Sadeghian, Antonello Rizzi
Neurocomputing, 170, 368–383, 2015.

An early contribution to data-driven modelling of complex power systems through dissimilarity learning and one-class Pattern Recognition.

Computational Intelligence · Smart Grids
DOI · IRIS

Short-Term Electric Load Forecasting Using Echo State Networks and PCA Decomposition

Filippo Maria Bianchi, Enrico De Santis, Antonello Rizzi, Alireza Sadeghian
IEEE Access, 3, 1931–1943, 2015.

Combines recurrent neural computation, dimensionality reduction, and evolutionary optimization for forecasting in a real-world electrical grid.

Recurrent Networks · Forecasting
DOI · IRIS

A Review of Solar and Wind Energy Forecasting: From Single-Site to Multi-Site Paradigm

Alessio Verdone, Massimo Panella, Enrico De Santis, Antonello Rizzi
Applied Energy, 392, 126016, 2025.

A critical overview of modern renewable-energy forecasting, with particular attention to the transition from local prediction to spatially distributed and multi-site learning.

Renewable Energy · Spatio-Temporal Learning
DOI · IRIS

Beyond Perplexity: A Multi-Faceted Analysis of a Novel Densely Connected Transformer

Enrico De Santis, Alessio Martino, Antonello Rizzi
Applied Sciences, 16(6), 2721, 2026.

Investigates densely connected Transformer architectures beyond perplexity, combining controlled architectural comparison, ablation, probing, and structural analysis of generated language through Zipf–RQA.

Transformer Architecture · Complexity · Structural Analysis
DOI · IRIS

2025: A GPT Odyssey. Deconstructing Intelligence by Gradual Dissolution of a Transformer

Enrico De Santis, Alessio Martino, Edoardo Bruno, Antonello Rizzi
2025 International Joint Conference on Neural Networks — IJCNN, 2025.

Investigates the hierarchical organization of Transformer representations through progressive GPT-2 layer ablation, connecting information granulation, semantic performance, and mechanistic understanding.

Transformer Hierarchy · Mechanistic Understanding
DOI · IRIS

LSTM in Recursive Feedback Loops: A Study on Textual Evolution and Complexity

Enrico De Santis, Alessio Martino, Francesca Ronci, Antonello Rizzi
2025 International Joint Conference on Neural Networks — IJCNN, 2025.

Studies recursive machine-generated text through complexity measures, revealing progressive structural degradation, model drift, and the erosion of long-range correlations.

Complexity · Generative Feedback
DOI · IRIS

Noosemia: Toward a Cognitive and Phenomenological Account of Intentionality Attribution in Human–Generative AI Interaction

Enrico De Santis, Antonello Rizzi
arXiv preprint arXiv:2508.02622, 2025. 

Introduces noosemia as a theoretical framework for investigating perceived agency, intentionality attribution, and the experience of artificial signs as manifestations of mind.

Human–AI Interaction · Cognition · Epistemology

For the complete and continuously updated publication record, see the bibliographic profiles below.