{"id":19278158,"url":"https://github.com/kotlarmilos/meta-features-anomaly-detection","last_synced_at":"2026-05-13T13:44:56.408Z","repository":{"id":53902934,"uuid":"269445272","full_name":"kotlarmilos/meta-features-anomaly-detection","owner":"kotlarmilos","description":"Set of meta-features for model selection in anomaly detection tasks based on domain-specific properties","archived":false,"fork":false,"pushed_at":"2022-01-31T21:48:09.000Z","size":16067,"stargazers_count":1,"open_issues_count":0,"forks_count":1,"subscribers_count":1,"default_branch":"master","last_synced_at":"2025-01-05T16:09:28.440Z","etag":null,"topics":["anomaly-detection","automl","machine-learning","meta-features"],"latest_commit_sha":null,"homepage":"","language":"Jupyter Notebook","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/kotlarmilos.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null}},"created_at":"2020-06-04T19:19:31.000Z","updated_at":"2023-10-09T08:39:31.000Z","dependencies_parsed_at":"2022-08-13T03:50:39.052Z","dependency_job_id":null,"html_url":"https://github.com/kotlarmilos/meta-features-anomaly-detection","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/kotlarmilos%2Fmeta-features-anomaly-detection","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/kotlarmilos%2Fmeta-features-anomaly-detection/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/kotlarmilos%2Fmeta-features-anomaly-detection/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/kotlarmilos%2Fmeta-features-anomaly-detection/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/kotlarmilos","download_url":"https://codeload.github.com/kotlarmilos/meta-features-anomaly-detection/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":240383162,"owners_count":19792771,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["anomaly-detection","automl","machine-learning","meta-features"],"created_at":"2024-11-09T21:08:32.033Z","updated_at":"2026-05-13T13:44:51.372Z","avatar_url":"https://github.com/kotlarmilos.png","language":"Jupyter Notebook","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Novel Meta-Features for Automated Machine Learning Model Selection in Anomaly Detection\n\nThis repository contains the scripts which evaluate a novel set of meta-features for model selection in anomaly detection tasks based on domain-specific properties. \n\nBy using different kinds of metadata, such as the properties of the data, algorithm properties, or\ncorrelation previously derived from the data, it is possible to select different models to effectively solve a\ngiven anomaly detection task. The meta-learning approach based on a set of meta-features that describes\ndata properties and correlation can enable efficient model selection in AutoML frameworks.\n\nExperiments with 63 datasets from different repositories with varying schemas show that\nthe proposed set of meta-features achieves the accuracy of 87% for model selection, while the achieved\naccuracy for simple meta-features is 74%, for statistical meta-features 68%, for information theory metafeature\n70%, and for a comprehensive set of meta-features by [pyMFE](https://pypi.org/project/pymfe/) 73%.\n\nResults are in [/results](https://github.com/kotlarmilos/meta-features-anomaly-detection/tree/master/pycharm/results) directory. Evaluated algorithms are in [/algorithms](https://github.com/kotlarmilos/meta-features-anomaly-detection/tree/master/pycharm/algorithms) directory.\n\n## Datasets\n\nDatasets are collected from repositories in [Harvard Dataverse](https://dataverse.harvard.edu/dataset.xhtml?persistentId=doi:10.7910/DVN/OPQMVF) and [Numenta Anomaly Benchmark](https://github.com/numenta/NAB) and cover a broad range of domains including manufacturing, transportation, healthcare, intrusion detection, and system log analysis.\n\nDataset repository is available [here](https://drive.google.com/drive/folders/1hp9ZuVQRnRduzKzMMnyNQ52Gv9QqQATP?usp=sharing) and [here](https://drive.google.com/drive/folders/10j-x2yJsBOSsjkeX1S8FtkaJqfrkkJ7-?usp=sharing).\n\n## Prerequests\n\n - Python 3.7\n - Pip\n - Numpy\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fkotlarmilos%2Fmeta-features-anomaly-detection","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fkotlarmilos%2Fmeta-features-anomaly-detection","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fkotlarmilos%2Fmeta-features-anomaly-detection/lists"}