Applying Neural Network Approach to Homomorphic Encrypted Data
| Autor: | Adrian Ionita, Valentin-Alexandru Vladuta, Victor-Valeriu Patriciu, Ana-Maria Ghimes |
|---|---|
| Rok vydání: | 2018 |
| Předmět: |
060201 languages & linguistics
business.industry Computer science Big data Homomorphic encryption Cloud computing 06 humanities and the arts 02 engineering and technology Computer security computer.software_genre Encryption Outsourcing Domain (software engineering) Public-key cryptography 0602 languages and literature 0202 electrical engineering electronic engineering information engineering 020201 artificial intelligence & image processing Confidentiality business computer |
| Zdroj: | 2018 10th International Conference on Electronics, Computers and Artificial Intelligence (ECAI). |
| DOI: | 10.1109/ecai.2018.8679085 |
| Popis: | Nowadays, a new era is beginning, the Data Era, which comes with a multitude of challenges and opportunities. Due to technology advancement and increased demand for powerful computing resources, many organizations outsource their storage and computing services to providers of such services. The benefits of cloud computing are numerous, but the security of cloud stored data remains an important issue. Cyber-attacks can be hard to detect and the culprits hard to identify, which encourages the cyber attackers to further pursue their malicious purposes. Big Data combined with analytical methods can produce the best defensive techniques against intrusions in the cyber security domain. Neural networks are not yet feasible to work over encrypted data, but with a well-designed architecture for such an analytical method they can contribute to the security of a system. Our paper aims to present a solution for the protection of personal data in insecure environments, which can also use the same data to perform analytical tasks without affecting confidentiality and security of individuals. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|