|
Title:
|
A METHODOLOGY FOR GENERATING SYNTHETIC PERSONAL INFORMATION REPOSITORIES |
|
Author(s):
|
Mousa Abu Kashef, Seyed Shahrestani and Mohammed Al-Zobbi |
|
ISBN:
|
978-989-8704-46-7 |
|
Editors:
|
Piet Kommers, Tomayess Issa, Adriana Backx Noronha Viana, Theodora Issa and Pedro Isaías |
|
Year:
|
2022 |
|
Edition:
|
Single |
|
Keywords:
|
Synthesis Repositories, Personal Information, Data Bias |
|
Type:
|
Full Paper |
|
First Page:
|
85 |
|
Last Page:
|
94 |
|
Language:
|
English |
|
Cover:
|
|
|
Full Contents:
|
click to dowload 
|
|
Paper Abstract:
|
Researchers who need to work with personal information face complex privacy-related challenges. In almost all situations, personal information must be adequately protected. In some cases, the research works may require the utilization of data from actual patients. However, privacy and confidentiality concerns can make using such data impossible or challenging. In addition, access to actual data can be affected by several factors, including cost, ethical issues, acquiring consent, and other approval requirements. Anonymization, pseudonymization, and de-identification are approaches that provide some privacy protection. Even so, applying these approaches may not always be possible or appropriate. These approaches are intended to omit sensitive information, such as the patient's first or last name, ethnicity, spoken language, eye colour, skin colour, and the like. Nevertheless, some studies may require the use of such data. For these situations, synthetic data resembling factual patient information can provide the required resources. However, the synthetic data must be free from bias as much as practicable. Data bias can lead to an erroneous depiction of the phenomenon under study or the population. This paper presents a framework for creating synthetic personal data repositories without referring to actual data. The created synthetic data should be free from both selection and cognitive biases. |
|
|
|
|
|
|
