A Deep Matrix Factorization-Based Anomaly Detection Framework
Kejian Li, Ping Deng, Dexian Wang
Xihua University Chengdu University of Traditional Chinese Medicine
阅读操作
确认中在文库中上传 PDF 后可生成中文音频讲解。
摘要与影响
The objective of anomaly detection is to identify data points in a dataset that deviate significantly from the normal distribution. However, existing algorithms have the limitation of being unable to effectively capture the nonlinear characteristics of the data, and their performance degrades when processing large-scale datasets. To address this challenges, this paper proposes a deep matrix factorization-based anomaly detection framework. This framework applies deep learning update rules to a matrix factorization framework, effectively improving the optimisation strategies of traditional matrix factorization framework. Furthermore, six anomaly detection algorithms are developed based on different activation functions. Experiments on eight datasets demonstrate that DMFAD delivers excellent performance.
逐年被引趋势
暂无年度引用数据
关键指标
同类平均 = 1
同领域 · 同年份 · 同类型
Google Scholar 与 OpenAlex 的被引统计范围不同,数值存在差异属正常。
AI 辅助阅读
依据:摘要
可就本文提问;依据不足时会说明。
学术脉络
学科主题
计算机 / AIAnomaly Detection Techniques and Applications
Software System Performance and Reliability · Fault Detection and Control Systems