Textual orContextual ETL — Learn free

Self-check

Questions 1–10 of 10

1. What does textual ETL do with raw text?
Question 1A. It reads text, disambiguates it and places the words found, together with their context, into a structured databaseB. It compresses documents to save storage spaceC. It translates text into other languagesD. It converts structured tables into free-form proseE. It deletes documents that contain no numbers
2. How does textual ETL know which words to pluck out of text?
Question 2A. It selects the most frequently occurring wordsB. By fiat of one or more taxonomies that flag which words are important and provide their contextC. A systems programmer hard-codes each wordD. It chooses words at randomE. It extracts only words that appear in capital letters
3. What is inline contextualisation in textual ETL?
Question 3A. Translating each word into a numeric codeB. Attaching document metadata to every rowC. Selecting a word and its context by its position in the raw text and its proximity to other words and phrasesD. Sorting words alphabetically before loadingE. Summarising a document into a single sentence
4. What is the key difference between structured data and textual data illustrated by the $5000 bank deposit and the phrase 'I like biscuits'?
Question 4A. Structured data is larger than textual dataB. Textual data is always older than structured dataC. Structured data is stored in files while text is stored in databasesD. Structured data is precise whereas textual data is probabilisticE. Textual data cannot be stored in a database
5. What is needed before analytics can be run across both structured data and textual data together?
Question 5A. A second data warehouseB. A licence for each documentC. A dedicated systems programmerD. Deleting the structured data firstE. Some common connector between the two types of data
6. What is the document card catalogue used for in a corporation?
Question 6A. Quickly and efficiently finding corporate documents, including documents that relate to each otherB. Recording employee attendanceC. Storing customer credit card detailsD. Listing which books employees have borrowedE. Logging server performance
7. Roughly what proportion of the data in a corporation is estimated to be text?
Question 7A. 5% to 10%B. 85% to 95%C. 25% to 35%D. 50% exactlyE. Less than 1%
8. In the real-estate example of textual ETL, how are the words 'owner' in one document and 'trustee' in another treated?
Question 8A. They are stored in separate databasesB. The word 'trustee' is discarded as a duplicateC. They are recorded as the same data because their logical meaning is the same, even though the text differsD. They are flagged as a conflict for a consultant to resolveE. Only the more frequent word is kept
9. What distinguishes the word column of a textual ETL database from the name column of a standard database?
Question 9A. The word column is always emptyB. The word column holds only numbersC. The word column can hold only a single word per documentD. The word column is heterogeneous, holding any kind of word found in the document, whereas the name column is homogeneousE. The word column is defined by metadata before any document is read
10. In a textual ETL database, what does the classification (or context) column contain?
Question 10A. The file size of the source documentB. The date on which the document was loadedC. The name of the analyst who ran the processD. The number of times a word appearsE. The classification that gives meaning to the word in the same row, such as 'flu' being a disease or 'Eliquis' a medication