《异常检测——从经典算法到深度学习》9 异常检测资料汇总（持续更新&抛砖引玉）

浅浅的花香味﹌ 2022-10-28 14:24 14阅读 0赞

# 《异常检测——从经典算法到深度学习》 #

*  [0 概论][0]
 *  [1 基于隔离森林的异常检测算法 ][1 _]
 *  [2 基于LOF的异常检测算法][2 _LOF]
 *  [3 基于One-Class SVM的异常检测算法][3 _One-Class SVM]
 *  [4 基于高斯概率密度异常检测算法][4]
 *  [5 Opprentice——异常检测经典算法最终篇][5 Opprentice]
 *  [6 基于重构概率的 VAE 异常检测][6 _ VAE]
 *  [7 基于条件VAE异常检测][7 _VAE]
 *  [8 Donut: 基于 VAE 的 Web 应用周期性 KPI 无监督异常检测][8 Donut_ _ VAE _ Web _ KPI]
 *  [9 异常检测资料汇总（持续更新&抛砖引玉）][9]
 *  [10 基于条件 VAE 的鲁棒无监督KPI异常检测][10 _ VAE _KPI]
 *  [11 针对大量出现的KPI流快速部署异常检测模型][11 _KPI]
 *  [12 对复杂 KPI 基于VAE对抗训练的非监督异常检测][12 _ KPI _VAE]

相关：

*  [ VAE 模型基本原理简单介绍 ][VAE _]
 *  [ GAN 数学原理简单介绍以及代码实践 ][GAN _]

# 9 异常检测资料汇总（持续更新&抛砖引玉） #

考虑到异常检测任务的繁琐，这里特地列举一些可能大家用的到的论文和资料，如果有补充请务必在后面留言。

## 9.1 论文 ##

论文是重中之重，请务必重视论文阅读。但是在这里很难也不可能列举出所有优秀的论文，所以这里只提供一些常用的查论文的方法等。

*  [AIOps 的论文收集][AIOps]
 *  [谷歌学术][Link 1]
 *  [百度学术][Link 2]
 *  [中国知网][Link 3]

<table> 
 <thead> 
  <tr> 
   <th align="center">论文名称</th> 
   <th align="center">年份</th> 
   <th align="center">标签</th> 
   <th align="center">下载地址</th> 
   <th align="center">源码</th> 
   <th align="center">个人笔记</th> 
  </tr> 
 </thead> 
 <tbody> 
  <tr> 
   <td align="center">Isolation Forest</td> 
   <td align="center">2009</td> 
   <td align="center">Isolation Forest/集成学习</td> 
   <td align="center"><a href="https://www.lanzoui.com/icum7xe" rel="nofollow">蓝奏云</a></td> 
   <td align="center">sklearn</td> 
   <td align="center"><a href="https://blog.csdn.net/smileyan9/article/details/106249594" rel="nofollow">CSDN1</a></td> 
  </tr> 
  <tr> 
   <td align="center">LOF: Identifying Density-Based Local Outliers</td> 
   <td align="center">2000</td> 
   <td align="center">基于分布密度差异/基于局部计算</td> 
   <td align="center"><a href="https://www.lanzoui.com/icvqkaj" rel="nofollow">蓝奏云</a></td> 
   <td align="center">sklearn</td> 
   <td align="center"><a href="https://blog.csdn.net/smileyan9/article/details/106296832" rel="nofollow">CSDN2</a></td> 
  </tr> 
  <tr> 
   <td align="center">Opprentice: Towards Practical and Automatic Anomaly Detection Through Machine Learning</td> 
   <td align="center">2015</td> 
   <td align="center">随机森林 / 集成学习 / KPI</td> 
   <td align="center"><a href="https://netman.aiops.org/wp-content/uploads/2015/11/liu_imc15_Opprentice.pdf" rel="nofollow">netman</a></td> 
   <td align="center">\</td> 
   <td align="center"><a href="https://blog.csdn.net/smileyan9/article/details/106587227" rel="nofollow">CSDN5</a></td> 
  </tr> 
  <tr> 
   <td align="center">Variational Autoencoder based Anomaly Detection using Reconstruction Probability</td> 
   <td align="center">2015</td> 
   <td align="center">VAE</td> 
   <td align="center"><a href="https://www.lanzoui.com/iUmwshknzmh" rel="nofollow">蓝奏云</a></td> 
   <td align="center">\</td> 
   <td align="center"><a href="https://blog.csdn.net/smileyan9/article/details/109255466" rel="nofollow">CSDN6</a></td> 
  </tr> 
  <tr> 
   <td align="center">Unsupervised Anomaly Detection via Variational Auto-Encoder for Seasonal KPIs in Web Applications</td> 
   <td align="center">2018</td> 
   <td align="center">VAE / KPI</td> 
   <td align="center"><a href="https://netman.aiops.org/wp-content/uploads/2018/03/www2018.pdf" rel="nofollow">netman</a></td> 
   <td align="center"><a href="https://github.com/NetManAIOps/donut" rel="nofollow">github</a></td> 
   <td align="center"><a href="https://smileyan.blog.csdn.net/article/details/112307506" rel="nofollow">CSDN8</a></td> 
  </tr> 
  <tr> 
   <td align="center">Robust and Unsupervised KPI Anomaly Detection Based on Conditional Variational Autoencoder</td> 
   <td align="center">2018</td> 
   <td align="center">cVAE / KPI</td> 
   <td align="center"><a href="https://netman.aiops.org/wp-content/uploads/2018/12/camera_ready.pdf" rel="nofollow">netman</a></td> 
   <td align="center"><a href="https://github.com/NetManAIOps/Bagel" rel="nofollow">github</a></td> 
   <td align="center"><a href="https://smileyan.blog.csdn.net/article/details/113463339" rel="nofollow">CSDN10</a></td> 
  </tr> 
  <tr> 
   <td align="center">Rapid Deployment of Anomaly Detection Models for Large Number of Emerging KPI Streams</td> 
   <td align="center">2018</td> 
   <td align="center">Framework/集群</td> 
   <td align="center"><a href="https://netman.aiops.org/wp-content/uploads/2018/12/bujiahao.pdf" rel="nofollow">netman</a></td> 
   <td align="center">\</td> 
   <td align="center"><a href="https://smileyan.blog.csdn.net/article/details/114763060" rel="nofollow">CSDN11</a></td> 
  </tr> 
  <tr> 
   <td align="center">Unsupervised Anomaly Detection for Intricate KPIs via Adversarial Training of VAE</td> 
   <td align="center">2019</td> 
   <td align="center">VAE/GAN</td> 
   <td align="center"><a href="https://netman.aiops.org/wp-content/uploads/2019/04/chenwenxiao_infocom2019.pdf" rel="nofollow">netman</a></td> 
   <td align="center">\</td> 
   <td align="center"><a href="https://smileyan.blog.csdn.net/article/details/116106826" rel="nofollow">CSDN12</a></td> 
  </tr> 
  <tr> 
   <td align="center">Anomaly Detection with Generative AdversarialNetworks for Multivariate Time Series</td> 
   <td align="center">2018</td> 
   <td align="center">GAN/RNN/LSTM</td> 
   <td align="center"><a href="https://arxiv.org/pdf/1809.04758.pdf" rel="nofollow">arxiv</a></td> 
   <td align="center"><a href="https://github.com/LiDan456/MAD-GANs" rel="nofollow">github</a></td> 
   <td align="center"><a href="https://smileyan.blog.csdn.net/article/details/118312582" rel="nofollow">CSDN13</a></td> 
  </tr> 
  <tr> 
   <td align="center">Robust Random Cut Forest Based Anomaly Detection On Streams</td> 
   <td align="center">2016</td> 
   <td align="center">随机森林/RRCF</td> 
   <td align="center"><a href="http://proceedings.mlr.press/v48/guha16.pdf" rel="nofollow">proceedings</a></td> 
   <td align="center"><a href="https://github.com/kLabUM/rrcf" rel="nofollow">github</a></td> 
   <td align="center">padding</td> 
  </tr> 
  <tr> 
   <td align="center">Practical and White-Box Anomaly Detection through Unsupervised and Active Learning</td> 
   <td align="center">2020</td> 
   <td align="center">随机森林/RRCF</td> 
   <td align="center"><a href="https://netman.aiops.org/wp-content/uploads/2020/08/ICCCN2020-YaoWang.pdf" rel="nofollow">netman</a></td> 
   <td align="center">\</td> 
   <td align="center">padding</td> 
  </tr> 
 </tbody> 
</table>

## 9.2 数据集 ##

这个是特别需要关注的地方，因为没有数据集就没法做实验，没有数据集是一个非常严重的问题。到目前为止能够直接使用的数据集也很少。

*  [智能运维比赛][Link 4]
 *  [阿里云智能运维算法大赛][Link 5]
 *  [华为网络AI学习赛][AI] 注：数据集不能直接下载
 *  [KDD CUP 99][]
 *  [CTF\_data][CTF_data]
 *  [OmniAnomaly][]

其他数据集：

*  [现已公开的数据集汇总][Link 6]
 *  [旋转机械故障诊断公开数据集汇总][Link 7]
 *  [odds][]
 *  [Intel Lab Data][]

## 9.3 学习资源 ##

华为云学院提供了微认证，推荐新手了解一下。

*  NAIE日志异常检测服务应用 https://edu.huaweicloud.com/certifications/78d55ca9c9fa4b40b73f7979ffab9a43
 *  NAIE KPI异常检测模型服务应用 https://edu.huaweicloud.com/certifications/3310e5169bf54940bf644d0201ab7013

## 9.4 其他资源 ##

一些源代码可以在github上找到。

*  [https://github.com/heyanyidui/MAD-GANs][https_github.com_heyanyidui_MAD-GANs]
 *  [https://github.com/NetManAIOps/OmniAnomaly][OmniAnomaly]
 *  [https://github.com/NetManAIOps/donut][https_github.com_NetManAIOps_donut]
 *  [https://github.com/iopsai/iops][https_github.com_iopsai_iops]

## 9.5 资源征集 ##

由于自己能力有限，收集到的资料也相当有限，抱歉！将会持续更新！

如果有任何觉得相关或者可以用来做异常检测算法实验的数据集，都可以留言分享一下。在这里代表需要的小伙伴们 **感谢分享！**

--------------------

编写不易，拒绝白piao。。。

**感谢** 您的 **阅读**、**点赞**、**收藏** 和 **评论** ，别忘了 还可以 **关注** 一下哈，**感谢** 您的支持！

--------------------

> Smileyan  
> 2021.3.7 21:57

[0]: https://blog.csdn.net/smileyan9/article/details/106239187
[1 _]: https://blog.csdn.net/smileyan9/article/details/106249594
[2 _LOF]: https://blog.csdn.net/smileyan9/article/details/106296832
[3 _One-Class SVM]: https://blog.csdn.net/smileyan9/article/details/106351651
[4]: https://smileyan.blog.csdn.net/article/details/106398233
[5 Opprentice]: https://blog.csdn.net/smileyan9/article/details/106587227
[6 _ VAE]: https://blog.csdn.net/smileyan9/article/details/109255466
[7 _VAE]: https://blog.csdn.net/smileyan9/article/details/109508502
[8 Donut_ _ VAE _ Web _ KPI]: https://blog.csdn.net/smileyan9/article/details/112307506
[9]: https://smileyan.blog.csdn.net/article/details/113728611
[10 _ VAE _KPI]: https://smileyan.blog.csdn.net/article/details/113463339
[11 _KPI]: https://smileyan.blog.csdn.net/article/details/114763060
[12 _ KPI _VAE]: https://smileyan.blog.csdn.net/article/details/116106826
[VAE _]: https://smileyan.blog.csdn.net/article/details/107362252
[GAN _]: https://smileyan.blog.csdn.net/article/details/115307274
[AIOps]: https://netman.aiops.org/publications/
[Link 1]: https://scholar.google.com/
[Link 2]: https://xueshu.baidu.com/
[Link 3]: https://www.cnki.net/
[Link 4]: http://iops.ai/
[Link 5]: https://tianchi.aliyun.com/competition/gameList/activeList
[AI]: https://www.hwtelcloud.com/competitions/home
[KDD CUP 99]: http://kdd.ics.uci.edu/databases/kddcup99/
[CTF_data]: https://github.com/NetManAIOps/CTF_data
[OmniAnomaly]: https://github.com/NetManAIOps/OmniAnomaly
[Link 6]: http://www.twistedwg.com/2018/04/09/download-dataset.html
[Link 7]: https://blog.csdn.net/abbott_zb/article/details/102563347
[odds]: http://odds.cs.stonybrook.edu/
[Intel Lab Data]: http://db.csail.mit.edu/labdata/labdata.html
[https_github.com_heyanyidui_MAD-GANs]: https://github.com/heyanyidui/MAD-GANs
[https_github.com_NetManAIOps_donut]: https://github.com/NetManAIOps/donut
[https_github.com_iopsai_iops]: https://github.com/iopsai/iops