Oct 03, 20248 min read

DATAMINING QUESTIONS

In which part of the CRISP methodology we perform the test design activity?

  • Data Understanding
  • Evaluation
  • Business Understanding
  • Modeling

Talking about the general idea of database, what is the purpose of the “Schema on write” strategy? Scegli una o più alternative:

  • Optimization for specific types of queries
  • Clean design of the data structure
  • Avoid preprocessing of data before writing
  • Flexibility and efficiency for any kind of query

Talking about the general idea of database, what is the meaning of “Schema on write”? Scegli un’alternativa:

  • Create a schema for data after writing into the database
  • Create a schema for data while writing into the database
  • Change the schema of the data after each write
  • Create a schema for data before writing into the database

Talking about the general idea of database, what is the purpose of the “Schema on read” strategy? Scegli una o più alternative:

  • Avoid preprocessing of data before writing
  • Optimization for various types of queries
  • Possibility to extract data in various shapes
  • Flexibility for any kind of query

Talking about the general idea of database, what is the meaning of “Schema on read”? Scegli un’alternativa:

  • Write the raw data in the database, then at reading time shape then according to the reader’s needs
  • Create a schema for data after reading them from the database
  • Read the schema of the data before each data read
  • Change the schema Of the data before each read

Which of the definition below describes the OLAP operation Roll-Up? Scegli un’alternativa:

  • Reduces data aggregation and adds a detail level to a hierarchy
  • Changes the layout, in order to analyse a group of data from a different viewpoint
  • Reduces the number of cube dimensions after setting one of the dimensions to a specific value
  • Creates a link between concepts in interrelated cubes, to compare them
  • Causes an increase in data aggregation and removes a detail level in a hierarchy

Which of the definition below describes the OLAP operation Pivot? Scegli un’alternativa:

  • Causes an increase in data aggregation and removes a detail level in a hierarchy
  • Reduces data aggregation and adds a detail level to a hierarchy
  • Changes the layout, in order to analyse a group of data from a different viewpoint
  • Creates a link between concepts in interrelated cubes, to compare them
  • Reduces the number of cube dimensions after setting one of the dimensions to a specific value

Which of the definition below describes the OLAP operation Drill-Down? Scegli un’alternativa:

  • Reduces data aggregation and adds a detail level to a hierarchy
  • Creates a link between concepts in interrelated cubes, to compare them
  • Changes the layout, in order to analyse a group of data from a different viewpoint
  • Causes an increase in data aggregation and removes a detail level in a hierarchy
  • Reduces the number of cube dimensions after setting one of the dimensions to a specific value
  • Usage of dictionaries to solve inconsistencies
  • Snapshot of the operational data
  • Denormalisation
  • Calculation of derived data
  • Elimination of duplicates
  • Association of a timestamp to the operational data
  • Usage of dictionaries to solve inconsistencies
  • Snapshot of the operational data
  • Elimination Of duplicates
  • Snapshot of the operational data
  • Association of a timestamp to the operational data
  • Usage Of dictionaries to solve inconsistencies

Order the ETL operations in order to obtain the correct sequence

    1. Cleansing
    1. Transformation
    1. Extraction
    1. Loading

Order: 3-1-2-4

Which of the texts describes the DFM schema below? Scegli un’alternativa:

  • We need to track the purchases with credit cards. Each purchase has an amount, an exchange and a credit limit. The purchases are done in a date and in a store of a given type; stores have also a country and an area. Purchases have a card type (e.g. credit or debit## and a card name (e.g. MyCard or JeansCard##. Card types have also an issuing bank and a country.
  • We need to track the purchases with credit cards. Each purchase has an amount, an exchange and a credit limit. The purchases are done in a date and in a store of a given type and country; stores have also an area. Purchases have a card name (e.g. MyCard or JeansCard##t card_names have a card type (e.g. credit or debit##. Card types have also an issuing bank and a country.
  • We need to track the purchases with credit cards. Each purchase has an amount, an exchange and a credit limit. The purchases are done in a date and in a store of a given type; stores types have also a country and an area. Purchases have a card type (e.g. credit or debit## and a card name (e.g. MyCard or JeansCard##. Card names have also an issuing bank and a country.

Which of the following are typical data warehouse queries? Scegli una o più alternative:

  • Which products maximize the profit?
  • Which is the stock of product “XXX” in the “YYY” warehouse
  • What is the total revenue of yesterday in shop “B003” ?
  • What is the total revenue per product category and state?
  • What is the relationship between profits gained by products “WWW” and “ZZZ”?

Select the sentences describing a characteristic of a Data Warehouse Scegli una o più alternative:

  • Solves the inconsistencies
  • Non volatile
  • Includes the time dimension
  • Constantly updated as soon as the base data are updated
  • Includes all the base data in their native format

Associate the various Data Warehouse Architectures to the respective advantages

    1. Single Layer: c. The occupation of space is minimised
    1. Two Layers: a. The workloads for analysis and daily operations are separated
    1. Three Layers: b. An intermediate level of consolidated data is available

Which of the activities below is part of “Business Understanding” in the CRISP methodology?

  • Which data are available?
  • Which machine learning functions are necessary for my problem?
  • Which data must be collected with a specific campaign?
  • Which are the resources available (manpower. hardware, software,…)

Rank the technologies for increasing level of abstraction: 1 = generate the most specific information, … 3 = higher level knowledge

  • SQL queries on operational databases → 1
  • Data Mining on operational databases → 3
  • OLAP analysis on Data Mart → 2

Which of the following sentences describes an advantage of a Data Warehouse with respect to a standard DBMS Scegli una o più alternative:

  • Allows efficient execution of key-based queries
  • Manages efficiently data updates
  • Has tools for helping to solve inconsistencies
  • Allows analysis along the time dimension
  • Allows efficient execution of multi-dimensional queries
  • Slice dice → 2. Reduces the number of cube dimensions after setting one of the dimensions to a specific value
  • Drill across → 1. Creates a link between concepts in interrelated cubes, in order to compare them
  • Pivot → 5. Changes the layout, in order to analyse a group of data from a different viewpoint
  • Drill down → 4. Reduces data aggregation and adds a detail level to a hierarchy
  • Roll-up → 3. Causes an increase in data aggregation and removes a detail level in a hierarchy

What is Data Ingestion?

  • A process that copies data from sources to a repository, taking care of possible differences in speed between the generation and the storing process
  • A process that copies data from sources to a repository, ensuring high data quality
  • A process that copies data from sources to a repository, making the transformation required by the users
  • A process that copies data from sources to a Data Warehouse guaranteeing the correctness of data with respect to the schema

Check the features of Data Lakes in comparison with Data Warehouses, Multiple answers allowed. Selecting a wrong option will be penalised.

  • Can run on less expensive HW/SW
  • Do not enforce data quality
  • More complex management
  • Completely defined use cases

Check the main objectives pursued when choosing a Data Lake architecture

  • Store the data directly from the sources, as they are; the most appropriate structure will be decided later, according to the user needs
  • The data must be accurately structured, in order to provide fast answers to complex queries
  • The storage must be scalable and cheap, at the expenses of latency
  • Ensure high quality of data through an accurate cleaning and transformation step

Graph View

Backlinks