Misinformation Detection

A Signal Processing and Machine Learning Perspective
Erscheinungsjahr 2026
ISBN: 978-3-032-41561-5
Verlag: Springer

Buch, Englisch, 101 Seiten, Format (B × H): 168 mm x 240 mm

Reihe: Synthesis Lectures on Signal Processing

In an era when misinformation spreads faster than ever, cutting-edge statistical and machine learning techniques offer a powerful defense. This book provides a rigorous, technical deep dive into misinformation detection, combining statistical modeling, signal processing, and machine learning to tackle false information across digital media.

Beginning with an overview of the role of statistics in detecting misinformation, the book explores latent variable models, independent vector analysis, semi-supervised learning, and multimodal fusion. In that way, the book offers advanced yet practical methodologies for real-world applications.

A dedicated section on system integration and deployment focuses on how these techniques can move beyond theory into actionable solutions, covering explainability, evaluation, and human-in-the-loop aspects. The book also critically examines the social, ethical, and financial implications of misinformation, and provides insights into policy challenges and freedom of expression concerns.

Zielgruppe

Professional/practitioner

Autoren/Hrsg.

Corizzo, Roberto

Boukouvalas, Zois

Casimiro Cavalcante, Charles

Fachgebiete

• Mathematik | Informatik EDV | Informatik Informatik Künstliche Intelligenz Maschinelles Lernen
• Mathematik | Informatik EDV | Informatik Informatik Bildsignalverarbeitung
• Technische Wissenschaften Elektronik | Nachrichtentechnik Nachrichten- und Kommunikationstechnik Signalverarbeitung
• Technische Wissenschaften Elektronik | Nachrichtentechnik Elektronik
• Mathematik | Informatik EDV | Informatik Informatik Künstliche Intelligenz Mustererkennung, Biometrik

Inhaltsverzeichnis

Chapter 1: Motivation and Background.- Chapter 2: Machine Learning and Signal Processing Methods for Misinformation Detection.- Chapter 3: Signal Processing in Misinformation Detection Systems.- Chapter 4: Integration into a System / Software Product.- Chapter 5: Broader Implications and Ethical Considerations.- Chapter 6: Open Issues, Future Research Directions, and Conclusion.

Über Autor(innen)

Roberto Corizzo  is an Associate Professor in the Department of Computer Science at American University. He develops efficient and adaptive machine learning methods that continuously learn from evolving data while remaining reliable in high-stakes applications. His research focuses on continual learning, forecasting, anomaly detection, explainable AI, and class-imbalanced problems, with applications in finance, medical healthcare, cybersecurity, social networks, astrophysics, environmental monitoring, and scientific discovery. He co-authored over 85 articles, including 30 publications in international journals such as IEEE Transactions on Neural Networks and Learning Systems, Neural Networks, and Machine Learning, and international conferences such as AAAI, ICDM, ECML/PKDD, and ECAI. He is an Associate Editor for international journals including Machine Learning Journal of Big Data and International Journal of Data Science and Analytics. He served as Area Chair for the ECML/PKDD conference and on the program committees of international conferences such as ICML, AAAI, KDD, NeurIPS, and ICDM, and reviewed for several international journals.

Zois Boukouvalas  is an Associate Professor in the Department of Mathematics and Statistics at American University, where he serves as Director of the M.S. programs in Data Science and Artificial Intelligence. He earned his Ph.D. in Applied Mathematics from the University of Maryland, Baltimore County (UMBC) in 2017 under the guidance of Dr. Tülay Adali.

His research focuses on developing reliable and responsible artificial intelligence, with particular emphasis on multimodal machine learning, trustworthy AI, and rigorous evaluation of AI systems. His work explores how information from multiple data sources and modalities can be integrated to build AI systems that are not only accurate, but also reliable, interpretable, and appropriate for real-world use.

His research spans applications in scientific discovery, natural language processing, biomedical data, chemistry, and materials science, including the use of AI to accelerate the discovery and characterization of novel materials. More broadly, his work seeks to bridge advances in machine learning with complex scientific and societal problems while addressing questions of reliability, transparency, robustness, and responsible deployment.

Charles Casimiro Cavalcante  is a Full Professor with the Teleinformatics Engineering Department from Federal University of Ceara, Brazil, holding the Statistical Signal Processing Chair. From August 2014 to July 2015, he was a Visiting Assistant Professor with the Department of Computer Science and Electrical Engineering (CSEE), University of Maryland Baltimore County (UMBC), USA and a Senior Visiting Professor at Université Côte d’Azur, Nice France, from December 2023 to February 2024.

His main research interests include statistical signal processing, machine learning,  signal processing for communications,  and information geometry. He has authored three international patents and has worked on several funded research projects in the signal processing and wireless communications areas. He is the coauthor of the book ``Unsupervised Signal Processing: Channel Equalization and Source Separation'' (CRC Press) and a co-editor of the book ``Signals and Images: Advances and Results in Speech, Estimation, Compression, Recognition, Filtering, and Processing'' (CRC Press). He is a Senior Member of the IEEE and of the Brazilian Telecommunications Society (SBrT).