Skip to content

Detecção de anomalias em tempo real para identificação de fraudes com streaming data e online learning: uma revisão da literatura

Oct 2026 · Zenodo (CERN European Organization for Nuclear Research)
Anomaly Detection Techniques and Applications

Abstract

Resumo A crescente digitalização dos serviços financeiros e das plataformas electrónicas tem intensificado os desafios associados à detecção de fraudes, exigindo sistemas capazes de analisar grandes volumes de dados em tempo real. Os métodos tradicionais, baseados em regras fixas e em modelos estáticos, revelam limitações perante a evolução contínua dos padrões fraudulentos. O presente artigo tem como objectivo analisar, através de uma revisão da literatura, a aplicação da detecção de anomalias em tempo real à identificação de fraudes com recurso a Streaming Data e a algoritmos de Online Learning. O estudo adopta uma abordagem qualitativa, descritiva e exploratória, assente na pesquisa bibliográfica em bases de dados científicas e na análise qualitativa de conteúdo de artigos, livros e estudos especializados nas áreas da Inteligência Artificial, da aprendizagem automática, da mineração de fluxos de dados e da segurança digital. Os resultados evidenciam que a integração entre o processamento contínuo de dados, a detecção de anomalias e a aprendizagem incremental permite desenvolver sistemas mais adaptativos, escaláveis e eficientes na identificação de comportamentos suspeitos. Persistem, contudo, desafios relacionados com o concept drift, o desequilíbrio entre classes, os falsos positivos, a interpretabilidade dos modelos e a validação destas tecnologias em contextos reais, sendo ainda escassos os estudos sobre países em desenvolvimento, incluindo Moçambique. Conclui-se que a combinação entre Streaming Data e Online Learning constitui uma abordagem promissora para reforçar os mecanismos de prevenção de fraudes digitais, recomendando-se o desenvolvimento de estudos empíricos que considerem diferentes contextos tecnológicos e socioeconómicos. Abstract The increasing digitalisation of financial services and electronic platforms has intensified the challenges associated with fraud detection, requiring systems capable of analysing large volumes of data in real time. Traditional methods, based on fixed rules and static models, show limitations in the face of the continuous evolution of fraudulent patterns. This article aims to analyse, through a literature review, the application of real-time anomaly detection to fraud identification using Streaming Data and Online Learning algorithms. The study adopts a qualitative, descriptive and exploratory approach, based on a bibliographic search of scientific databases and on a qualitative content analysis of articles, books and specialised studies in the fields of Artificial Intelligence, machine learning, data stream mining and digital security. The findings show that integrating continuous data processing, anomaly detection and incremental learning enables the development of more adaptive, scalable and efficient systems for identifying suspicious behaviour. However, challenges persist regarding concept drift, class imbalance, false positives, model interpretability and the validation of these technologies in real-world settings, and studies on developing countries, including Mozambique, remain scarce. The study concludes that combining Streaming Data and Online Learning is a promising approach to strengthening digital fraud prevention mechanisms, and recommends empirical studies that consider different technological and socio-economic contexts.

View source

Similar papers

#artificial intelligence Open access May 2023

Evaluating the Performance of Large Language Models on GAOKAO Benchmark

GAOKAO-Bench is introduced, an intuitive benchmark that employs questions from the Chinese GAOKAO examination as test samples, including both subjective and objective questions that contribute a robust evaluation benchmark for future large language models and offers valuable insights into the advantages and limitations...

Xiaotian Zhang, Chun-yan Li, Yi Zong et al. · 216 citations · ⚡17
#artificial intelligence Open access Jul 2024

Gender, Race, and Intersectional Bias in Resume Screening via Language Model Retrieval

This work investigates the possibilities of using LLMs in a resume screening setting via a document retrieval framework that simulates job candidate selection and finds that the MTEs are biased, significantly favoring White-associated names in 85% of cases and female-associated names in only 11.1% of cases.

Kyra Wilson, Aylin Caliskan · 131 citations · ⚡8
#artificial intelligence Review Nov 2024

How to Build a Quantum Supercomputer: Scaling from Hundreds to Millions of Qubits

This work shows that orders of magnitude enhancement in performance could be obtained by a combination of hardware improvements and tight quantum-HPC integration and introduces high-performance architectures for quantum-probabilistic computing with custom-designed accelerators to tackle today's industry-scale classical...

Masoud Mohseni, Artur Scherer, K. Johnson et al. · 121 citations · ⚡9
#artificial intelligence Review Oct 2025

Ultralytics YOLO Evolution: An Overview of YOLO26, YOLO11, YOLOv8 and YOLOv5 Object Detectors for Computer Vision and Pattern Recognition

This paper presents a comprehensive overview of the Ultralytics YOLO family, emphasizing architectural evolution, benchmarking, deployment, and emerging directions from YOLOv5 through YOLO27, and examines detection, segmentation, depth, classification, pose, oriented detection, tracking, export, quantization, and deplo...

Ranjan Sapkota, Manoj Karkee · 112 citations · ⚡10

The Death of Schema Linking? Text-to-SQL in the Age of Well-Reasoned Language Models

This work revisits schema linking when using the latest generation of large language models (LLMs) and finds empirically that newer models are adept at utilizing relevant schema elements during generation even in the presence of large numbers of irrelevant ones.

Karime Maamari, Fadhil Abubaker, Daniel Jaroslawicz et al. · 109 citations · ⚡19

BadRAG: Identifying Vulnerabilities in Retrieval Augmented Generation of Large Language Models

A novel threat is unveiled in which attackers steer the RAG system's response by injecting malicious passages into its knowledge base, enabling the attacker to steer the response without altering the user input or modifying the RAG weights.

Jiaqi Xue, Meng Zheng, Yebowen Hu et al. · 109 citations · ⚡8

Related blog posts

MIT News · Artificial Intelligence Sep 29, 2026

Who we become when we talk to machines

Professor Sherry Turkle’s new book, “Artificial Intimacy,” offers a withering critique of chatbots and the antisocial dynamics she believes they encourage.

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.