"Bird Cloud" is a cloud computing brand under Shenzhen Qianhai Bird Cloud Computing Co., Ltd., and a leading enterprise-level cloud computing service provider in China. The team has many years of industry experience, focusing on the research and development of cloud computing technology, providing all-round cloud computing solutions based on intelligent cloud servers for developers, government and enterprise users and financial institutions, and providing users with reliable enterprise-level public cloud services.
The frequent data breaches every year always bring some lessons, one of which is that it is never too late to take data protection measures. Fortunately, companies are paying more attention to data privacy, and big data is one of their most concerned areas.
Just yesterday, five former employees of Microsoft said in an interview with Reuters that Microsoft's vulnerability report data had been illegally invaded in 213, but this incident was not exposed at that time.
Former employees of Microsoft said that it took Microsoft more than a month to fix all the security vulnerabilities listed in the hacked database, so the leaked vulnerability information would not have much impact on users of Windows products. At that time, Microsoft also hired a third-party company to investigate the incident to find out whether there were attackers using the leaked vulnerability information to launch attacks on the network, but the company did not find any attacks related to related vulnerabilities.
Mary Shacklett is the president of Transworld Data, a technology research and market development company. As an insider, she gives some suggestions to enterprise management to ensure that their big data adopts reliable data privacy practices.
one way to achieve anonymity is to encrypt personally identifiable data elements. Another method is to identify the data of individuals with similar value, and then average it into a comprehensive income value, which will be integrated into larger data analysis. Other methods include data revision or masking.
collecting digital information generated by government, enterprises and individuals creates great opportunities for decision-making based on knowledge and information. Driven by mutual benefit, data can be exchanged and released between parties in need. However, in its original form, data usually contains sensitive personal information, and publishing these data will violate personal privacy. Privacy protection under aggregate data publishing is an important and challenging problem. Most of the existing technologies use generalization and holistic deletion methods, and we propose a partial (partial) deletion method to anonymize aggregate data. This method ensures that no matter how much prior knowledge the attacker has, the strong association rules about sensitive information will no longer appear in the anonymized data. This method not only greatly reduces the information loss, but also provides the choice of useful association rules that tend to keep the distribution of original data or protect mining according to the requirements of downstream usage scenarios. The preliminary evaluation shows that our method is more than 1 times better than other methods in maintaining the original data distribution, retaining more useful association rules and introducing only a few false rules, and reducing the information loss by about 3% on average.
The above is only a part of the work on data privacy. There are more ways to protect data privacy, such as determining the departments involved in big data within the company and regularly reviewing the data privacy of these departments. Finally, when formulating and implementing data privacy protection measures, it needs to be based on the business needs and development of enterprises.