Data Imputation — Learn free

Self-check

Questions 1–10 of 10

1. What does the word Imputation mean in the context of data?
Question 1A. Removing duplicate recordsB. Filling in missing dataC. Encrypting sensitive valuesD. Sorting data into categoriesE. Converting text into numbers
2. In which domain is Data Imputation majorly used?
Question 2A. Data GovernanceB. Database AdministrationC. Network SecurityD. Data ScienceE. Data Archiving
3. Why does Data Imputation normally come into play?
Question 3A. So business rules can be imposed on all data rather than on subsets of dataB. So data can be stored more compactlyC. So sensitive values can be masked before sharingD. So historical data can be retired safelyE. So queries can run faster on large clusters
4. Before a Machine Learning model runs, which two ways of handling missing data are described for Data Imputation?
Question 4A. Encrypting the missing values or flagging them for reviewB. Duplicating the row or moving it to an archive tableC. Converting the value to zero or converting it to textD. Splitting the dataset or merging it with anotherE. Filling in the missing data or removing the whole row
5. Which kinds of activity are said to give imperfect results when data is missing?
Question 5A. Backing up, restoring and mirroringB. Compressing, archiving and indexingC. Categorising, classifying and clusteringD. Encrypting, hashing and signingE. Sorting, filtering and paging
6. In numeric datasets, what value is generally used to replace missing values?
Question 6A. The MEAN valueB. ZeroC. The maximum valueD. The most recent valueE. A randomly generated value
7. In the first imputation technique described, what happens to data that is missing at random?
Question 7A. It is replaced with the median of the columnB. It is flagged and kept in the datasetC. It is estimated from a separate reference datasetD. It is completely removed from the dataset before useE. It is replaced with the previous row's value
8. How does the second imputation technique treat values that are missing at random?
Question 8A. They are left blank for the model to handleB. They are replaced by referring to other data points available within the current datasetC. They are replaced with values from an external systemD. They are replaced with zeroE. They cause the entire dataset to be rejected
9. The third imputation technique applies when data is NOT missing at random. How are those values mainly replaced?
Question 9A. By removing every affected columnB. By copying values from the nearest rowC. By asking the data owner to supply themD. By replacing them with null markersE. Using Mean, Median or Mode values
10. Which of these is NOT described as a way of dealing with missing data in Data Imputation?
Question 10A. Removing the whole row that contains the missing valueB. Replacing the missing value with the MEANC. Leaving the missing values in place and training the model directly on themD. Replacing the value by referring to other data points in the datasetE. Replacing the value with the Median or Mode