Bull Sas

Commissariat a L'energie Atomique Et Aux Energies Alternatives

Kaoutar Sghiouer

Pierre Seroul

Jean-François Roy

Guillaume Porcher

Techniques for selecting a learning model defined in particular by parameters and hyperparameters from among a plurality of learning models, implemented by a computing device and provided. A computing device may be provided includes a model selection module and a model repository including a plurality of series of instructions each corresponding to a learning model and each including hyperparameter values. The method may be provided that includes a step of selecting a model when the prediction performance value and the classification value are greater than a predetermined second threshold and the hyperparameter value is greater than a predefined threshold value.

Method and system for selecting a learning model from among a plurality of learning models

The invention relates to a data processing method for preparing a dataset that includes a processor that receives a first plurality of data input streams to prepare an output dataset. The plurality of data input streams and the output dataset are different. The method includes standardizing the plurality of data input streams, encoding the normalized data, preprocessing missing data, and transmitting a preprocessed dataset. The invention further relates to a data processing system, and a recording medium on which the data processing program is recorded.

Automatic dataset preprocessing

The invention relates to a device and a method () for determining a technical incident risk value in an infrastructure (), said method comprising:

Method and device for determining a technical incident risk value in a computing infrastructure from performance indicator values

The invention relates to a method and device for determining an anomaly prediction performance index value, said method comprising:

Method and device for determining a performance indicator value for predicting anomalies in a computing infrastructure from values of performance indicators

A method for determining an estimated duration before a technical incident in a computing infrastructure, executed by a computing device. The computing device including a data processing module, a storage module that stores in memory at least one correlation base between performance indicators, wherein the correlation base includes values of duration before becoming anomalous between correlated performance indicators, and a collection module. The method includes receiving performance indicator values, identifying anomalous performance indicators, identifying first and other at-risk indicators. The method includes determining an estimated duration before a technical incident including a calculation, from the anomalous indicators and at-risk indicators, a shorter path leading to a risk of technical incident, and a calculation of an estimated duration before a technical incident. The estimated duration is calculated from the values of duration before becoming anomalous between correlated performance indicators for each of the performance indicators constituting the shortest path calculated.

Growing community of inventors

Compiegne, France

Kaoutar Sghiouer