At the Institute of Information Technology of ANAS, the next seminar devoted to the theme "Classification of large data by protecting privacy" was held.

The leading researcher of the institute, PhD in technical sciences, associate professor Fargana Abdullayeva noted the relevance of the use of in-depth training methods in the protection of confidentiality of personal data in Big data time series.

The emergence of Big data with the widespread use of "smart" devices and the fact that the data, behavior, transcripts, social media information, and health information in the records were become more urgent.

According to the speaker, even though data privacy and data security terms are used as synonyms, data security ensures that data is only accessible to authorized people and guarantees the legitimate use of the data and ensures data privacy protection by Data anonymization.

The classic methods used to protect sensitive data anonymity encounter considerable problems in detecting sensitive information in Big data time series. It is suggested that the use of in-depth training approaches to protect the confidentiality of personal data is used to address these difficulties, she noted.

Mrs. Abdullayeva highlighted the method of in-depth training based on avtoencoder proposed to carry out the analysis of data by protecting the privacy of personal data: The main purpose of the process is to find sensitive information within the individual's personal data and to transform them into non-sensitive information. To implement this process, the architecture consisted of two blocks. The blocks of architecture are composed of two modular "sparse denoising auto-encoders" and two neural networks called the Convolutional Neural Network ("Transformed data classification function").

© All rights reserved. Citing to www.ict.az is necessary upon using news