s-X-AIPI self-X open-source toolset in R
self-X autonomic abilities for data mining activities
![s-X-AIPI self-X open-source toolset in R](/sites/default/files/styles/16_9_100/public/2024-06/s-X-AIPI_Rtoolset.png?itok=XQdIurNU)
self-X autonomic abilities for data mining activities
"selfX" R package includes the implementation of several functions to perform data exploration activities (as feature selection) and model training (as algorithm selection or anomaly detection) in a more autonomic way. Besides, complementary functions for data cleaning, statistical tests and outlier and missing values treatment have been included.
This R package is implemented as open-source software within s-X-AIPI project and offered to the R community for use and improvement.
The s-X-AIPI project has received funding from the European Union's Horizon Europe research and innovation programme under grant agreement Nº 101058715 (s-X-AIPI)
There are multiple methodologies proposed for the development of AI pipelines in industrial environments, and all of them include a "Data Exploration" phase in which the treatment of outlier data, the analysis of correlations and feature extraction/selection (among others) are carried out. Furthermore, "Model Training" is also crucial to detect anomalies and make predictions in order to obtain valuable insights for make informed decision-making.
With the aim of making data mining activities more autonomic, some ad-hoc procedure were defined as part of the Asphalt Use Case within the s-X-AIPI project and subsequently adapted for general use.
Main approaches included in the R package are:
- self-X autonomic supervised and unsupervised feature selection
- self-X autonomic unsupervised anomaly detection
- self-X autonomic predictive algorithm selection
- self-X autonomic data preparation: cleaning data, statistical tests or outliers treatment activities