yzhao062 / anomaly-detection-resources Goto Github PK

提交一本关于路由器异常检测的书籍
Anomaly-Detection and Health-Analysis Techniques for Core Router Systems
https://link.springer.com/book/10.1007%2F978-3-030-33664-6#toc

如题，感觉中文版本与英文版本区别不大，仅导语部分有区别。
是不是之后可以做更多的一些本地化的操作？

Reference 83: https://federation.edu.au/__data/assets/pdf_file/0011/443666/ICDM2018-Tutorial-Final.pdf is broken.

I have tried some outlier detection datasets (ODDs) in this website like Annthyroid dataset (http://odds.cs.stonybrook.edu/annthyroid-dataset/).

However, when I compare some ordinary supervised models (e.g., SVM and Random Forest), the results indicate that SVM and RF are much better than the anomaly detection algorithms like OC-SVM and Isolation Forest.

I was wonder the reason for this weird results, because threoratically the outlier detection algorithms should perform better in the outlier detection task. Could anyone help me figure this problem? Thanks!

https://developer.ibm.com/learningpaths/get-started-anomaly-detection-api/what-is-anomaly-detection/?msclkid=1ea92091b56311ec8b5c4ae53ec234a0

I googled outlier detection time series and I found this appeared first:

A review on outlier/anomaly detection in time series data

This review was composed this year. I have read half of it and I think it is informative.

Will you consider to append it to the time series table?

Hi,

We have developed a novel approach for detecting outliers in categorical datasets which you might want to add to this excellent repository.

• Theory: https://onlinelibrary.wiley.com/doi/full/10.1111/sjos.12407
• Software: https://github.com/mlindsk/molic

/Mads

It always turns out:'This preview took too long to generate.
But you can view the raw file.'

我想要做一个异常检测系统，但准确率、召回不高

Paper: https://arxiv.org/abs/1802.04431

Open source code: https://github.com/khundman/telemanom

Title: SSD: A Unified Framework for Self-Supervised Outlier Detection
Link: https://openreview.net/forum?id=v5gjXpmR8J (ICLR 2021)
Code: https://github.com/inspire-group/SSD

I cannot open to link to the XGBOD paper.

the following is the error message. I guess I just dont have access right to this url.

Secure Connection Failed

An error occurred during a connection to www.yuezhao.me. PR_END_OF_FILE_ERROR

The page you are trying to view cannot be shown because the authenticity of the received data could not be verified.
Please contact the website owners to inform them of this problem.

Currently, the implementation of SimpleDetectorAggregator does not allow for novelty detection usages.

The method _create_scores(self, X) does apply standardization based on the tensor of scores X. The data evaluated for novelty detection are not transformed the same way as the data used to fit the SimpleDetectorAggregator, and thus the threshold defined at fitting can not be applied to determine wether or not the data is a novelty.

One notable consequence is that running SimpleDetectorAggregator(...).predict(X[0, :]) (novelty detection on a single point) will always output the a score of 0.

SimpleDetectorAggregator should instead keep the scalers used when fitting and use them to process the data when creating new scores. This would add support for novelty detection.

谢啦

Hi Yue,

I found your repository is amazing. Just let you know that our research group (the inventor of iForest) recently has proposed new isolation based anomaly detection methods that you may like to include:

1. Ting, Kai Ming, Bi-Cun Xu, Takashi Washio, and Zhi-Hua Zhou. "Isolation Distributional Kernel: A New Tool for Kernel based Anomaly Detection." In Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, pp. 198-206. 2020.

code: https://github.com/IsolationKernel/Codes/tree/main/IDK

2. Bandaragoda, Tharindu R., Kai Ming Ting, David Albrecht, Fei Tony Liu, Ye Zhu, and Jonathan R. Wells. "Isolation‐based anomaly detection using nearest‐neighbor ensembles." Computational Intelligence 34, no. 4 (2018): 968-998.

code: https://github.com/zhuye88/iNNE

Cheers,
Ye Zhu

119 R packages for outlier detection are described at https://github.com/pridiltal/ctv-AnomalyDetection

https://github.com/hrbrmstr/AnomalyDetection/blob/master/README.md
Anomaly Detection Using Seasonal Hybrid Extreme Studentized Deviate Test

Twitterfolks launched this package in 2014. Many coding and package standards have changed. The package now conforms to CRAN standards.

The plots were nice and all but terribly unnecessary. The two core functions have been modified to only return tidy data frames (tibbles, actually). This makes it easier to chain them without having to deal with list element dereferencing.

Shorter, snake-case aliases have also been provided:

ad_ts for AnomalyDetectionTs
ad_vec for AnomalyDetectionVec

The original names are still in the package but the README and examples all use the newer, shorter versions.

The following outstanding PRs from the original repo are included:

Added in PR #98 (@gggodhwani)
Added in PR #93 (@nujnimka)
Added in PR #69 (@randakar)
Added in PR #44 (@nicolasmiller)
PR #92 (@caijun) inherently resolved

If those authors find this repo, please add yourselves to the DESCRIPTION as contirbutors.

感觉更像是采用了随机失活所抛弃掉的方法，尽管这样说可能有点刻薄了。另外预训练那一块论文也没详细说明

This is one of the most comprehensive work towards the true definition of anomaly detection. I think you'd want to include it.

https://www.groundai.com/project/a-meta-analysis-of-the-anomaly-detection-problem/

See here
https://slideplayer.com/slide/6437518/

Very cool, informative site !
I am currently looking for benchmark tests for anomaly detections in Big Data. Do you know anything about this?