Detecting Data Anomalies with Python in Metaverse: IQR, MAD, LOF, and KNN Models Explored (XRP/USD Coin Sample)
DeFi Adoption in the Energy and Utilities Industry: Toward Sustainable Digital Transformation, Apple Acad Press Inc, ss.77-100, 2026
- Yayın Türü: Kitapta Bölüm / Araştırma Kitabı
- Basım Tarihi: 2026
- Doi Numarası: 10.1201/9781779642400-4
- Yayınevi: Apple Acad Press Inc
- Sayfa Sayıları: ss.77-100
- Kırklareli Üniversitesi Adresli: Evet
Özet
This chapter explores the critical role of data polishing and anomaly detection in enabling decentralized finance (DeFi)-driven digital transformation within the energy and utilities industry, with broader implications for dataintensive environments such as cryptocurrency markets and metaverse ecosystems. In an ideal digital infrastructure, decision-making systems operate on transparent, consistent, and high-quality data that support reliable automation, decentralized governance, and predictive analytics. Such an ecosystem presumes seamless data integrity, adaptive risk monitoring, and trustworthy financial and operational exchanges. In practice, however, industrial and financial platforms remain vulnerable to noisy datasets, measurement errors, systemic inconsistencies, and undetected anomalies, which undermine analytical accuracy and institutional confidence. Prior studies on machine learning, data cleaning, and blockchain-based energy systems emphasize preprocessing, normalization, and outlier detection as prerequisites for intelligent operations. Parallel research on crypto-market anomalies and metaverse security highlights the relevance of statistical and learning-based surveillance models. Yet, these strands often remain methodologically fragmented, rarely examining their integrated function within DeFi-enabled infrastructures. This chapter addresses this gap by advancing a unified analytical framework grounded in data reliability theory and decentralized analytics. Focusing on the comparative evaluation of IQR, MAD, and LOF models applied to XRP/USD datasets since 2018, the paper assesses robustness, sensitivity, and computational efficiency. The findings demonstrate how systematic data polishing strengthens trustless financial architectures, enhances operational resilience, and supports sustainable digital transformation in energy and utility ecosystems.