{"id":1428,"date":"2021-12-01T17:13:06","date_gmt":"2021-12-01T17:13:06","guid":{"rendered":"https:\/\/sappan-project.eu\/?p=1428"},"modified":"2021-12-01T17:15:00","modified_gmt":"2021-12-01T17:15:00","slug":"datasets-quality-assessment-for-machine-learning","status":"publish","type":"post","link":"https:\/\/sappan-project.eu\/?p=1428","title":{"rendered":"Datasets Quality Assessment For Machine Learning"},"content":{"rendered":"\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t\t
\n\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t
By Dominik Soukup (CESNET)<\/div>\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t

Have you ever heard about Machine learning (ML)? Probably yes, ML is a popular technique for network traffic classification and incident detection. However, have you ever heard about evaluating the quality of datasets (QoD)? \n
QoD is becoming more important with deployment ML in production, and project SAPPAN contributes to this topic.<\/h4>\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t\t
\n\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t

\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n<\/p>

The main prerequisite\nfor any ML model is the input dataset that is used for training. If the input\ndataset contains mistakes and irrelevant data, the ML model cannot work\ncorrectly too. Moreover, the network traffic domain is a dynamic environment\nwhere patterns can change dynamically over time, and each network is also\ndifferent. Therefore we need to take care of our datasets, measure their\nquality and improve them if necessary.<\/span><\/span><\/p>\n\n

 <\/span><\/span><\/p>\n\n

Currently, many\nauthors publish public datasets but with limited descriptions. Therefore, we\nneed to primarily trust the author’s reputation instead of verifying its\nquality for our network. The decision if we can use the available public\ndataset or create our own is repeated anytime we want to use the ML algorithm.\nThis decision is not easy since there is no standardized way to evaluate the\nQoD.<\/span><\/span><\/p>\n\n

 <\/span><\/span><\/p>\n\n

Our recent study [1]\nis focused on initial definitions of dataset quality and proposes a novel\nframework to assess dataset quality. We introduce the following definitions\nthat we are missing for standardized description of dataset quality: Good\nDataset, Better Dataset, and Minimal Dataset. In Figure 1. you can see the high\nlevel architecture of our novel framework that we used for experiments to\nmeasure datasets quality.<\/span><\/span><\/p>\n\n\n\n\n\n