Data Generalization/Blurring and Specialization

Self-check

Questions 1–10 of 10

1. Which analogy describes Data Generalisation versus Data Specialisation?
Question 1A. Encrypt versus decryptB. Backup versus restoreC. Read versus writeD. Zoom-out versus zoom-in, or less precise versus preciseE. Batch versus real-time
2. What does Data Generalisation do to data?
Question 2A. Summarises it in such a way that its usefulness remains intactB. Deletes all personal information permanentlyC. Splits each entity into several smaller entitiesD. Encrypts it with a private keyE. Duplicates it across several servers
3. What does Data Specialisation do?
Question 3A. Summarises several entities into oneB. Replaces values with dummy valuesC. Divides an entity into multiple entities to further make sense of the dataD. Removes identifying columnsE. Converts data into age ranges
4. Which of the following is the example of Data Generalisation given in I Am Datapedia!?
Question 4A. Splitting an account table into current and savings account tablesB. Creating a summarised Person entity that hosts both Students and Patients detailsC. Swapping ages between recordsD. Masking a credit card numberE. Replacing a password with asterisks
5. Which of the following is the example of Data Specialisation given in I Am Datapedia!?
Question 5A. Combining Students and Patients into a Person entityB. Grouping ages into ranges such as 20-29C. Hiding an email addressD. Merging two customer tables into oneE. Dividing an account table into separate current account and savings account tables
6. When individual ages are replaced by ranges such as 20-29 (2) and 30-39 (3), what is achieved?
Question 6A. The data becomes reversible to its original valuesB. The data loses all analytical valueC. The essence of the data is maintained but the original data is not exposedD. Each person can still be identified by exact ageE. The data is encrypted
7. What is the limitation of Declarative Generalisation?
Question 7A. It can only be done on a small dataset, because the data is distorted manuallyB. It cannot be applied to personal dataC. It requires a machine learning modelD. It is not accepted by regulatory bodiesE. It always removes the identifying columns
8. How does Automated Generalisation work?
Question 8A. Data is distorted by hand using manual queriesB. Data is deleted once it reaches an age thresholdC. Data is copied to a secure serverD. Data is split into multiple tablesE. Algorithms distort the data until it meets accepted standards
9. What is a Quasi Identifier in Data Generalisation?
Question 9A. A single column such as ID that uniquely identifies an entityB. A combination of multiple columns that together can identify an entityC. A column that has been maskedD. A randomly generated tokenE. A column defined by a regulatory body
10. Why might masking a single Direct Identifier such as Name fail to fully achieve Data Generalisation?
Question 10A. Because names cannot be maskedB. Because Direct Identifiers are never used for identificationC. Because masking always removes the whole recordD. Because a single identifier is not guaranteed to be unique, for example there can be several people named Hube in the same city or postcodeE. Because regulators forbid masking names